{"id":1507812,"date":"2022-04-06T11:24:16","date_gmt":"2022-04-06T18:24:16","guid":{"rendered":"https:\/\/www.esri.com\/arcgis-blog\/?post_type=blog&p=1507812"},"modified":"2022-04-20T09:11:16","modified_gmt":"2022-04-20T16:11:16","slug":"regression-analysis-when-the-data-doesnt-conform","status":"publish","type":"blog","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform","title":{"rendered":"Regression analysis: when the data doesn\u2019t conform"},"author":294792,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"open","ping_status":"closed","template":"","format":"standard","meta":{"_acf_changed":false,"_searchwp_excluded":""},"categories":[23341],"tags":[466322,37561],"industry":[],"product":[36801],"class_list":["post-1507812","blog","type-blog","status-publish","format-standard","hentry","category-analytics","tag-arcgis-insights","tag-regression-analysis","product-insights"],"acf":{"short_description":"A guided walkthrough of regression analysis in ArcGIS Insights","flexible_content":[{"acf_fc_layout":"content","content":"

Regression analysis<\/a> in ArcGIS Insights allows you to create a regression model using relationships between a dependant variable and one or more explanatory variables, then use that model to predict values. The regression analysis method that Insights uses is called Ordinary Least Squares (OLS), which is a linear regression method.<\/span>\u00a0<\/span><\/p>\n

One notable thing about linear regression is that it is designed specifically for normally distributed data. There are distinct rules and guidelines that should be followed to create a solid regression model; your variables should be normally distributed, there should be a linear relationship between your dependant and explanatory variables but no collinearity, you want an R2<\/sup> value close to one, little to zero skewness, the excess kurtosis close to 0, etc. However, as you know it can be challenging to meet such specific requirements when working with your real-world data. Data is rarely perfect, like in the case we are about to go through, so here you will look at one example of how you can work with imperfect data.<\/span>\u00a0<\/span><\/p>\n

This blog will use data information collected from buoys on the Great Lakes of North America. It will go through step by step how to explore and choose appropriate variables, create and evaluate a regression model, and use that model to predict variables in related datasets.<\/span>\u00a0<\/span><\/p>\n

 <\/p>\n"},{"acf_fc_layout":"content","content":"

Step 1: Open Insights and add data<\/span>\u00a0<\/span><\/h2>\n

In this first section you will retrieve your data, import it into Insights and prepare it for your regression analysis. If you are more familiar with ArcGIS Insights and this process, you can skip ahead to Step 2.<\/span>\u00a0<\/span><\/p>\n

    \n
  1. Navigate to <\/span>https:\/\/www.glahf.org\/data\/.<\/span><\/a><\/li>\n
  2. Scroll to <\/span>Mechanical Energy<\/span><\/b> section and click <\/span>download buoy locations & summaries<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"sidebar","content":"
      \n
    1. Open <\/span>ArcGIS Insights Desktop<\/span><\/b>.You can also use your Insights in ArcGIS Online or Insights in ArcGIS Enterprise.
      \n<\/span>
      \n<\/span>If necessary, accept the <\/span>Activation complete<\/span><\/b> message and skip the <\/span>Welcome to Insights<\/span><\/b> window.<\/span>
      \n<\/span>
      \n<\/span><\/li>\n
    2. Navigate to the <\/span>Datasets <\/span><\/b>tab on the home page.<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1532592,"id":1532592,"title":"Datasets Tab","filename":"Datasets-Tab-1.jpg","filesize":14050,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/datasets-tab-2","alt":"Image highlighting the Datasets tab in insights","author":"294792","description":"","caption":"","name":"datasets-tab-2","status":"inherit","uploaded_to":1507812,"date":"2022-03-25 18:20:52","modified":"2022-03-25 18:21:21","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":301,"height":598,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1.jpg","medium-width":131,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1.jpg","medium_large-width":301,"medium_large-height":598,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1.jpg","large-width":301,"large-height":598,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1.jpg","1536x1536-width":301,"1536x1536-height":598,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1.jpg","2048x2048-width":301,"2048x2048-height":598,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1-234x465.jpg","card_image-width":234,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Datasets-Tab-1.jpg","wide_image-width":301,"wide_image-height":598}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"
        \n
      1. Click <\/span>New dataset<\/span><\/b> and select the downloaded <\/span>buoy_data.gdp<\/span><\/b>.<\/span><\/span><\/span><\/li>\n<\/ol>\n

        Note: <\/strong>if using Insights in ArcGIS Online or ArcGIS Enterprise, you will need to update the Type<\/strong> to File geodatabase before adding.
        \n<\/strong><\/p>\n

          \n
        1. Download the workbook package<\/a>.<\/li>\n
        2. In ArcGIS Insights go to the Workbooks<\/span><\/b> tab and click <\/span>Import<\/span><\/b>.<\/span><\/span><\/li>\n
        3. Import the attached workbook and open it.<\/li>\n<\/ol>\n

          The Add to page<\/strong> dialogue should open automatically.<\/p>\n

            \n
          1. Go to the Local content <\/span><\/b>tab and select <\/span>buoy_data.gdb<\/span><\/b><\/b><\/li>\n
          2. Add buoy_locations_NOAA<\/span><\/b> and <\/span>mean_monthly_buoy_stats<\/span><\/b> from <\/span>bouy_data.gdb<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1540652,"id":1540652,"title":"Add to page","filename":"Add-to-page.jpg","filesize":61124,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/add-to-page","alt":"ArcGIS Insights Add to page dialogue with datasets from step 11 selected","author":"294792","description":"","caption":"","name":"add-to-page","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 17:18:50","modified":"2022-04-01 17:19:25","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1444,"height":807,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page.jpg","medium-width":464,"medium-height":259,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page.jpg","medium_large-width":768,"medium_large-height":429,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page.jpg","large-width":1444,"large-height":807,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page.jpg","1536x1536-width":1444,"1536x1536-height":807,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page.jpg","2048x2048-width":1444,"2048x2048-height":807,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page-826x462.jpg","card_image-width":826,"card_image-height":462,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Add-to-page.jpg","wide_image-width":1444,"wide_image-height":807}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"
              \n
            1. Delete the map from the page.<\/span><\/span><\/li>\n
            2. Click the D<\/span><\/b>ataset options<\/span><\/b> button on <\/span>mean_monthly_buoy_stats_noaa<\/span><\/b>.\u00a0<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

              \n"},{"acf_fc_layout":"image","image":{"ID":1540672,"id":1540672,"title":"Dataset Options","filename":"Dataset-Options.jpg","filesize":21320,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/dataset-options","alt":"Dataset options button","author":"294792","description":"","caption":"","name":"dataset-options","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 17:22:29","modified":"2022-04-01 17:22:56","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":661,"height":403,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options.jpg","medium-width":428,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options.jpg","medium_large-width":661,"medium_large-height":403,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options.jpg","large-width":661,"large-height":403,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options.jpg","1536x1536-width":661,"1536x1536-height":403,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options.jpg","2048x2048-width":661,"2048x2048-height":403,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options.jpg","card_image-width":661,"card_image-height":403,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Dataset-Options.jpg","wide_image-width":661,"wide_image-height":403}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                \n
              1. Click <\/span>Enable location<\/span><\/b>, switch to the <\/span>Geography <\/span><\/b>tab, and click <\/span>Run.<\/span><\/b>\u00a0<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1540662,"id":1540662,"title":"Enable location by geography","filename":"Enable-location-by-geography.jpg","filesize":27030,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/enable-location-by-geography","alt":"Enable location by geography tool settings","author":"294792","description":"","caption":"","name":"enable-location-by-geography","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 17:21:01","modified":"2022-04-01 17:23:59","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":704,"height":498,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography.jpg","medium-width":369,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography.jpg","medium_large-width":704,"medium_large-height":498,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography.jpg","large-width":704,"large-height":498,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography.jpg","1536x1536-width":704,"1536x1536-height":498,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography.jpg","2048x2048-width":704,"2048x2048-height":498,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography-657x465.jpg","card_image-width":657,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/03\/Enable-location-by-geography.jpg","wide_image-width":704,"wide_image-height":498}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                Many of the numeric variables are imported as strings. These fields can be converted to numbers in the data table. <\/span>The data originally retrieved from NOAA (National Oceanic and Atmospheric) uses 9999 as a no data replacement. As you convert the data from a string to a number field you can also remove the 9999 field to just convert the values you want<\/span>\u00a0<\/span><\/p>\n

                  \n
                1. Click the <\/span>Dataset options<\/span><\/b> button then choose <\/span>View data<\/span><\/b> table<\/span><\/b> from the menu.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1540742,"id":1540742,"title":"View data table","filename":"View-data-table.jpg","filesize":18018,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/view-data-table","alt":"Dataset options menu highlighting the View data table option","author":"294792","description":"","caption":"","name":"view-data-table","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 18:04:57","modified":"2022-04-01 18:05:27","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":663,"height":332,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table.jpg","medium-width":464,"medium-height":232,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table.jpg","medium_large-width":663,"medium_large-height":332,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table.jpg","large-width":663,"large-height":332,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table.jpg","1536x1536-width":663,"1536x1536-height":332,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table.jpg","2048x2048-width":663,"2048x2048-height":332,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table.jpg","card_image-width":663,"card_image-height":332,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/View-data-table.jpg","wide_image-width":663,"wide_image-height":332}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"
                    \n
                  1. Add a new field and enter <\/span>the expression<\/span> IF(AirTemp_mean <> ‘9999’, VALUE(AirTemp_mean))<\/span><\/b>.<\/span> Click Run.<\/span><\/span><\/li>\n
                  2. Rename your new field to <\/span>AirTemp<\/span><\/b><\/b>.<\/li>\n
                  3. Repeat step 15 for <\/span>WtrTemp_Mean<\/span><\/b> and <\/span>AtmoPress_Mean<\/span><\/b><\/b>.<\/li>\n
                  4. Rename your fields to <\/span>WaterTemp <\/span><\/b>and <\/span>AtmoPressure<\/span><\/b>.<\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1540762,"id":1540762,"title":"Data table calculations","filename":"Data-table-calculations.jpg","filesize":88337,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/data-table-calculations","alt":"Data table showing the newly calculated fields with the expression","author":"294792","description":"","caption":"","name":"data-table-calculations","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 18:21:30","modified":"2022-04-01 18:21:57","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1042,"height":655,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations.jpg","medium-width":415,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations.jpg","medium_large-width":768,"medium_large-height":483,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations.jpg","large-width":1042,"large-height":655,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations.jpg","1536x1536-width":1042,"1536x1536-height":655,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations.jpg","2048x2048-width":1042,"2048x2048-height":655,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations-740x465.jpg","card_image-width":740,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Data-table-calculations.jpg","wide_image-width":1042,"wide_image-height":655}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                    Now that the data is converted to the correct type, while there are no \u201c9999\u201d values there are now many rows with no data. You can use an advanced filter to filter out the null data.<\/span>\u00a0<\/span><\/p>\n

                      \n
                    1. Close the data table.<\/span><\/li>\n
                    2. In the data pane next to <\/span>mean_monthly_buoy_stats.table<\/span><\/b> click the <\/span>D<\/span><\/b>ataset options<\/span><\/b> button then choose <\/span>Advanced filter<\/span><\/b> from the menu.<\/span><\/li>\n
                    3. Enter the expression <\/span>AND(ISNOTNULL (AirTemp), ISNOTNULL( WaterTemp), ISNOTNULL(AtmoPressure))<\/span><\/b>.<\/span><\/li>\n
                    4. Click <\/span>Apply<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1540862,"id":1540862,"title":"Advanced filter","filename":"Advanced-filter.jpg","filesize":32163,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/advanced-filter","alt":"Advanced filter dialogue with xustom filter expression","author":"294792","description":"","caption":"","name":"advanced-filter","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:25:19","modified":"2022-04-01 19:25:47","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":978,"height":409,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter.jpg","medium-width":464,"medium-height":194,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter.jpg","medium_large-width":768,"medium_large-height":321,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter.jpg","large-width":978,"large-height":409,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter.jpg","1536x1536-width":978,"1536x1536-height":409,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter.jpg","2048x2048-width":978,"2048x2048-height":409,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter-826x345.jpg","card_image-width":826,"card_image-height":345,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Advanced-filter.jpg","wide_image-width":978,"wide_image-height":409}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                      Step 2: Explore variables\u00a0<\/span><\/h2>\n

                      Now that you have your data, the next step is to explore that data and determine which variables will be used for your regression model. If you skipped section one, please download this workbook<\/a>, import it into your ArcGIS Insights <\/span>Workbooks<\/span><\/b> tab and open it to begin with section 2.<\/span>\u00a0<\/span><\/p>\n

                        \n
                      1. Expand the mean_monthly_buoy_stats data set.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                        For the regression model you are going to need a dependant variable and one or more explanatory variables. The dependant variable will be the field you want to explain with your model and the explanatory variables will be used to explain that variable. The goal is to make a model to help predict water temperatures, so you know <\/span>WaterTemp<\/span><\/b> will be our dependent variable. To help determine the explanatory variables you can use scatter plots and histograms.<\/span>\u00a0<\/span><\/p>\n

                          \n
                        1. Select your newly created numeric fields <\/span>WaterTemp<\/span><\/b>, <\/span>AirTemp<\/span><\/b>, and\u00a0 <\/span>AtmoPressure <\/span><\/b>and drag them to the scatter plot matrix drop zone.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                          There must be a linear relationship between our dependant variable and our explanatory variable. For the relationship to be linear when one variable changes the other must change in the same proportion. However, you do not want collinearity, meaning a linear relationship between explanatory variables.\u00a0<\/span>\u00a0<\/span><\/p>\n

                          On the scatter plot matrix, <\/span>you can visually analyze if there is a linear relationship and look at the R2<\/sup> values. Visually a linear relationship will create a straight line when graphed. When examining your scatter plot it is also important to check for outliers. Outliers will stand apart from the predominant pattern of the graph and could be erroneous measures or a once in a lifetime event that would skew your results. Sometimes removing these outliers prior to continuing can show a higher linearity and produce a better model.\u00a0<\/span>\u00a0<\/span><\/p>\n

                          The R2<\/sup> value measures the strength of the relationship. You want the R2<\/sup> value to be close to 1, indicating a stronger relationship the closer it is to 1.\u00a0<\/span>\u00a0<\/span><\/p>\n

                          Looking at the scatter plot matrix created, you can immediately see that there is a linear relationship between Water Temp and Air Temp and confirm the relationship is strong from the high R2<\/sup> value. In this case there only seems to be one suitable explanatory variable; however, if there were more than one, you would want to ensure there was not a linear relationship between explanatory variables.<\/span>\u00a0<\/span><\/p>\n"},{"acf_fc_layout":"image","image":{"ID":1540872,"id":1540872,"title":"Scatter plot matrix","filename":"Scatter-plot-matrix.jpg","filesize":36589,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/scatter-plot-matrix","alt":"Scatter plot matrix using AirTemp, AtmoPress, and WateraTemp.","author":"294792","description":"","caption":"","name":"scatter-plot-matrix","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:27:41","modified":"2022-04-01 19:28:20","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":699,"height":672,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix.jpg","medium-width":271,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix.jpg","medium_large-width":699,"medium_large-height":672,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix.jpg","large-width":699,"large-height":672,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix.jpg","1536x1536-width":699,"1536x1536-height":672,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix.jpg","2048x2048-width":699,"2048x2048-height":672,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix-484x465.jpg","card_image-width":484,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Scatter-plot-matrix.jpg","wide_image-width":699,"wide_image-height":672}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                          Another factor to consider is that you want the variables to have a normal distribution. You can evaluate the distribution using a histogram.<\/span>\u00a0<\/span><\/p>\n

                            \n
                          1. Select the <\/span>AirTemp<\/span><\/b> field, drag it to the <\/span>Chart<\/span><\/b> drop zone, and drop it on <\/span>Histogram<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                            You can see off the bat this variable is not perfectly normal. To further confirm this, you can add the normal distribution curve to the histogram.<\/span>\u00a0<\/span><\/p>\n

                              \n
                            1. Click on the chart to activate it. Click the <\/span>Chart statistics<\/span><\/b> button then check the box next to <\/span>Normal distribution<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                              You can also evaluate how skewed the variable is \u2013 we can do this by looking at the mean and median values automatically placed on the chart. In this case the mean and median are equal, meaning the data has low skew.<\/span>\u00a0<\/span><\/p>\n"},{"acf_fc_layout":"image","image":{"ID":1540892,"id":1540892,"title":"Histogram with normal distribution curve","filename":"Histogram-with-normal-distribution-curve.jpg","filesize":34866,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/histogram-with-normal-distribution-curve","alt":"Histogram of AirTemp field with Mean, Median and Normal distribution curve.","author":"294792","description":"","caption":"","name":"histogram-with-normal-distribution-curve","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:29:43","modified":"2022-04-01 19:30:12","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":835,"height":530,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve.jpg","medium-width":411,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve.jpg","medium_large-width":768,"medium_large-height":487,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve.jpg","large-width":835,"large-height":530,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve.jpg","1536x1536-width":835,"1536x1536-height":530,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve.jpg","2048x2048-width":835,"2048x2048-height":530,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve-733x465.jpg","card_image-width":733,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-with-normal-distribution-curve.jpg","wide_image-width":835,"wide_image-height":530}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                              Another place we can look for statistics is on the back of the card.<\/p>\n

                                \n
                              1. Click the Flip card<\/strong> button.<\/li>\n<\/ol>\n

                                Here there are several statistics, including Skewness<\/strong> and Excess kurtosis<\/strong>. We want the skewness to be close to zero and the excess kurtosis to be close to 0.<\/p>\n

                                Looking at this variable we can see it is not perfectly normally distributed. We can try calculating the log of the variable to see if it creates a better normal distribution.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":1540902,"id":1540902,"title":"Histogram statistics","filename":"Histogram-statistics.jpg","filesize":22341,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/histogram-statistics","alt":"The statistics on the back of the histogram card highlighting the Skewness and Excess kurtosis variables","author":"294792","description":"","caption":"","name":"histogram-statistics","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:31:18","modified":"2022-04-01 19:31:50","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":512,"height":511,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics.jpg","medium-width":262,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics.jpg","medium_large-width":512,"medium_large-height":511,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics.jpg","large-width":512,"large-height":511,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics.jpg","1536x1536-width":512,"1536x1536-height":511,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics.jpg","2048x2048-width":512,"2048x2048-height":511,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics-466x465.jpg","card_image-width":466,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Histogram-statistics.jpg","wide_image-width":512,"wide_image-height":511}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                                  \n
                                1. Flip the card back over.<\/span><\/li>\n
                                2. Click the <\/span>Dataset options<\/span><\/b> button and choose <\/span>View data table<\/span><\/b>.<\/span><\/li>\n
                                3. Click <\/span>+Field<\/span><\/b>.<\/span><\/li>\n
                                4. Enter the function <\/span>log(AirTemp)<\/span><\/b> and click <\/span>R<\/span><\/b>un<\/span><\/b>.<\/span><\/li>\n
                                5. Rename the field to <\/span>Log AirTemp<\/span><\/b>.<\/span><\/li>\n
                                6. Close the data table.<\/span><\/li>\n
                                7. From the data pane select the Log AirTemp<\/strong> field and create a histogram.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                  In this case the log transformation actually makes the distribution worse. Having done this, you can now make the decision \u2013 while the AirTemp variable is not perfect it is not skewed and will be suitable enough to try creating a regression model.<\/span>\u00a0<\/span><\/p>\n"},{"acf_fc_layout":"content","content":"

                                  Step 3: Create and evaluate models<\/span>\u00a0<\/span><\/h2>\n

                                  Now that you have evaluated the variables you want to use in the regression model, this next stage will guide you through creating, evaluating, and comparing your regression models.<\/span>\u00a0<\/span><\/p>\n

                                    \n
                                  1. Select the <\/span>mean_monthly_buoy_stats<\/span><\/b> dataset and drag it to the <\/span>Model Creation<\/span><\/b> page.\u00a0<\/span><\/li>\n
                                  2. Expand the dataset and select the <\/span>WaterTemp<\/span><\/b> numeric field and drag it to a map drop zone.<\/span><\/li>\n
                                  3. Click on the map to activate it and click the <\/span>Action<\/span><\/b> button.<\/span><\/li>\n
                                  4. Go to the <\/span>Find Answers<\/span><\/b> tab and choose <\/span>How is it related?<\/span><\/b>\u00a0<\/span><\/li>\n
                                  5. Select <\/span>Regression Model<\/span><\/b>.<\/span><\/li>\n
                                  6. For <\/span>Choose a dependant variable<\/span><\/b>, select <\/span>WaterTemp<\/span><\/b>.<\/span><\/li>\n
                                  7. For <\/span>Choose explanatory variables<\/span><\/b>, check <\/span>AirTemp<\/span><\/b> and click <\/span>Select<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                    Note:<\/strong> you can click the <\/span>Visualize<\/span><\/b> button and a scatter plot will be created \u2013 you have already evaluated the scatter plot in the previous step so will skip this for now.<\/span>\u00a0<\/span><\/p>\n"},{"acf_fc_layout":"image","image":{"ID":1540922,"id":1540922,"title":"Regression Model settings","filename":"Regression-Model-settings.jpg","filesize":21760,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/regression-model-settings","alt":"Regression model tool dialogue with settings as described in steps 6 and 7","author":"294792","description":"","caption":"","name":"regression-model-settings","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:34:47","modified":"2022-04-01 19:35:20","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":335,"height":536,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings.jpg","medium-width":163,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings.jpg","medium_large-width":335,"medium_large-height":536,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings.jpg","large-width":335,"large-height":536,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings.jpg","1536x1536-width":335,"1536x1536-height":536,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings.jpg","2048x2048-width":335,"2048x2048-height":536,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings-291x465.jpg","card_image-width":291,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-Model-settings.jpg","wide_image-width":335,"wide_image-height":536}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                                      \n
                                    1. Click <\/span>Run<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                      A couple of things will happen when you run your model; a new layer <\/span>Avg Standardized Residual<\/span><\/b> will be added to your map, a new dataset <\/span>Predicted WaterTemp 1<\/span><\/b> will be added to the data pane and a Regression Model will be added to the data pane<\/span>\u00a0<\/span><\/p>\n

                                        \n
                                      1. Click on the layer symbol for the <\/span>Sum of WaterTemp<\/span><\/b> layer to hide this field from the map.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1540942,"id":1540942,"title":"Map with avg standardized residual","filename":"Map-with-avg-standardized-residual.jpg","filesize":28956,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/map-with-avg-standardized-residual","alt":"Map with avg standardized residualk and indicator of where to click to hide the pre-existing Sum of WaterTemp layer","author":"294792","description":"","caption":"","name":"map-with-avg-standardized-residual","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:38:51","modified":"2022-04-01 19:39:25","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":531,"height":532,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual.jpg","medium-width":261,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual.jpg","medium_large-width":531,"medium_large-height":532,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual.jpg","large-width":531,"large-height":532,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual.jpg","1536x1536-width":531,"1536x1536-height":532,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual.jpg","2048x2048-width":531,"2048x2048-height":532,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual-464x465.jpg","card_image-width":464,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Map-with-avg-standardized-residual.jpg","wide_image-width":531,"wide_image-height":532}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"
                                          \n
                                        1. Expand the <\/span>Predicted WaterTemp 1 <\/span><\/b>dataset. This dataset has all the same variables as your original but with 3 new ones: <\/span>Estimated<\/span><\/b>, <\/span>Residual<\/span><\/b>, and <\/span>Standardized Residual<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                          Note:<\/strong> You want the residuals to be normally distributed with a mean of zero. If this is not the case, then the coefficients\u2019 p-values are unreliable. You can check this in the same way we checked the normal distribution of our explanatory variable in Section 2, steps 3-5.\u00a0<\/span>\u00a0<\/span><\/p>\n

                                            \n
                                          1. Collapse the dataset.<\/span><\/li>\n
                                          2. Expand <\/span>Regression Model<\/span><\/b> 1<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                            Here the details of your model can be seen, and you can evaluate the results of your model to examine how good of a model it is. Looking at the adjusted R2<\/sup> you can see its close to one at 0.77. The Durban-Watson Test value which describes model autocorrelation is a little low, you want this value to be between 1.5 and 2.5, and the p-value of the model is 0 which is ideal.<\/span>\u00a0<\/span><\/p>\n"},{"acf_fc_layout":"image","image":{"ID":1540952,"id":1540952,"title":"Regression model 1 output","filename":"Regression-model-output.jpg","filesize":23815,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/regression-model-output","alt":"Regression model 1 details","author":"294792","description":"","caption":"","name":"regression-model-output","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:40:38","modified":"2022-04-01 19:41:00","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":367,"height":506,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output.jpg","medium-width":189,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output.jpg","medium_large-width":367,"medium_large-height":506,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output.jpg","large-width":367,"large-height":506,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output.jpg","1536x1536-width":367,"1536x1536-height":506,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output.jpg","2048x2048-width":367,"2048x2048-height":506,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output-337x465.jpg","card_image-width":337,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-output.jpg","wide_image-width":367,"wide_image-height":506}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                                            The last component is the options to view confidence intervals.<\/span>\u00a0<\/span><\/p>\n

                                              \n
                                            1. Click on <\/span>View confidence intervals<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                              This chart shows the level of confidence you have in the model \u2013 this is a point chart which automatically sets up confidence interval bars showing both the upper and lower 90, 95, and 99 percent standardized confidence intervals. This is especially valuable when comparing models. Now create a second regression model and compare.<\/span>\u00a0<\/span><\/p>\n

                                                \n
                                              1. Click the <\/span>A<\/span><\/b>ction<\/span><\/b> button<\/span><\/b> on the map and reopen <\/span>R<\/span><\/b>egression <\/span><\/b>M<\/span><\/b>odel<\/span><\/b> (under <\/span>Find answers<\/span><\/b> and <\/span>How<\/span><\/b> is it related?<\/span><\/b>).<\/span><\/li>\n
                                              2. For <\/span>Choose a <\/span><\/b>d<\/span><\/b>ependant<\/span><\/b> variable<\/span><\/b>, select <\/span>WaterTemp<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                                This time you will use the log air temp variable as the explanatory variable.<\/span>\u00a0<\/span><\/p>\n

                                                  \n
                                                1. For <\/span>Choose explanatory variables<\/span><\/b>, select <\/span>Log AirTemp<\/span><\/b> and click <\/span>Select<\/span><\/b>. Click <\/span>Run<\/span><\/b> to create the model.<\/span><\/li>\n
                                                2. In the data pane, expand Regression Model 2<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                                  Looking at the R2<\/sup> it is a little lower, but the Durban Watson is more in the range you would want it. Another way we can compare the models is to compare the confidence intervals on the same point chart.<\/span>\u00a0<\/span><\/p>\n"},{"acf_fc_layout":"image","image":{"ID":1540962,"id":1540962,"title":"Regression model 2 output","filename":"Regression-model-2-output.jpg","filesize":24447,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/regression-model-2-output","alt":"Regression model 2 details","author":"294792","description":"","caption":"","name":"regression-model-2-output","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:42:08","modified":"2022-04-01 19:42:20","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":370,"height":515,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output.jpg","medium-width":188,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output.jpg","medium_large-width":370,"medium_large-height":515,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output.jpg","large-width":370,"large-height":515,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output.jpg","1536x1536-width":370,"1536x1536-height":515,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output.jpg","2048x2048-width":370,"2048x2048-height":515,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output-334x465.jpg","card_image-width":334,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regression-model-2-output.jpg","wide_image-width":370,"wide_image-height":515}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                                                    \n
                                                  1. Select Regression Model 2 in the data pane and drag it to the existing point chart.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                                    You should see a second set added to the chart. The confidence interval bars of the new model are slightly longer \u2013 meaning you have less confidence in the output of that model. This leads me to decide the first model is more accurate and therefore the one you will want to use going forward!<\/span>\u00a0<\/span><\/p>\n"},{"acf_fc_layout":"image","image":{"ID":1540972,"id":1540972,"title":"Confidence intervals","filename":"Confidence-intervals.jpg","filesize":11332,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/confidence-intervals","alt":"Confidence intervals chart comparing both models","author":"294792","description":"","caption":"","name":"confidence-intervals","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:42:56","modified":"2022-04-01 19:43:12","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":512,"height":511,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals.jpg","medium-width":262,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals.jpg","medium_large-width":512,"medium_large-height":511,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals.jpg","large-width":512,"large-height":511,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals.jpg","1536x1536-width":512,"1536x1536-height":511,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals.jpg","2048x2048-width":512,"2048x2048-height":511,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals-466x465.jpg","card_image-width":466,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Confidence-intervals.jpg","wide_image-width":512,"wide_image-height":511}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                                                    Step 4: Predict variables<\/span>\u00a0<\/span><\/h2>\n

                                                    In this stage you will use the Regression model that was created in the previous step to predict variables in another dataset where data is missing.<\/span>\u00a0<\/span><\/p>\n

                                                      \n
                                                    1. Drag <\/span>Regression Model 1<\/span><\/b> to the <\/span>Predict Variables <\/span><\/b>page tab<\/span>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                                      This page should now have <\/span>Regression Model 1<\/span><\/b> and <\/span>BuoyData 2017<\/span><\/b> in the data pane and a map on the page.<\/span>\u00a0<\/span><\/p>\n

                                                      The 2017 buoy data here has a variety of information from buoys in the great lakes but no water temperature data. You are going to use the model you just created to try and predict the water temperatures!<\/span>\u00a0<\/span><\/p>\n

                                                        \n
                                                      1. Click on the Regression Model 1 and drag it onto the map and drop it in the <\/span>Predict Variable<\/span><\/b> drop zone.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                                        The Predict Variable dialog should open next to the map.<\/span>\u00a0<\/span><\/p>\n

                                                          \n
                                                        1. For <\/span>Choose <\/span><\/b>a<\/span><\/b> layer<\/span><\/b>, verify that <\/span>BuoyData 2017<\/span><\/b> is selected.<\/span><\/li>\n
                                                        2. For <\/span>Choose the<\/span><\/b> regression model<\/span><\/b> layer<\/span><\/b>, verify that <\/span>Regression Model 1 <\/span><\/b>is selected. The details of the model should appear below the parameter.<\/span><\/li>\n
                                                        3. Scroll down in the dialog and update the map variables. Select <\/span>air temp<\/span><\/b> as the Replacement field<\/strong>.<\/span><\/li>\n
                                                        4. Click <\/span>Run<\/span><\/b>.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

                                                          Now you have a new <\/span>Predicted Variable 1<\/span><\/b> dataset which has our Estimated water temperature value! The map has also been updated to show the average estimated water temperature. You can also use the estimated variable in other charts or tables to get a closer look at the variables.<\/span>\u00a0<\/span><\/p>\n

                                                            \n
                                                          1. Change the <\/span>month<\/span><\/b> field in the Predicted Variable 1 dataset from a number to a string by clicking on the icon to the left of the field name.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1541002,"id":1541002,"title":"Convert month field to string","filename":"Convert-month-field-to-string-1.jpg","filesize":5572,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/convert-month-field-to-string-2","alt":"Highlights the icon next to the month field and the options for converting that field to other field types including Number, Dtring, and Rate\/Ratio","author":"294792","description":"","caption":"","name":"convert-month-field-to-string-2","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 19:48:48","modified":"2022-04-01 19:49:29","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":297,"height":126,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1-213x126.jpg","thumbnail-width":213,"thumbnail-height":126,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1.jpg","medium-width":297,"medium-height":126,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1.jpg","medium_large-width":297,"medium_large-height":126,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1.jpg","large-width":297,"large-height":126,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1.jpg","1536x1536-width":297,"1536x1536-height":126,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1.jpg","2048x2048-width":297,"2048x2048-height":126,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1.jpg","card_image-width":297,"card_image-height":126,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Convert-month-field-to-string-1.jpg","wide_image-width":297,"wide_image-height":126}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"
                                                              \n
                                                            1. From the Predicted Variable 1 dataset select <\/span>station_code<\/span><\/b>, <\/span>month<\/span><\/b>, and <\/span>Estimated,<\/span><\/b> drag them to the Table drop zone, and drop on <\/span>Summary Table<\/span><\/b>.<\/span><\/li>\n
                                                            2. Change the summary statistic on the <\/span>Estimated<\/span><\/b> field to <\/span>AVG.<\/span><\/b><\/li>\n
                                                            3. Resize and re-style as desired!<\/span>\u00a0<\/span><\/li>\n<\/ol>\n"},{"acf_fc_layout":"image","image":{"ID":1541062,"id":1541062,"title":"Regressionmodel final output","filename":"Regressionmodel-final-output.jpg","filesize":54729,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output.jpg","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\/regressionmodel-final-output","alt":"Final map and table output of the regression model","author":"294792","description":"","caption":"","name":"regressionmodel-final-output","status":"inherit","uploaded_to":1507812,"date":"2022-04-01 20:24:31","modified":"2022-04-01 20:24:52","menu_order":0,"mime_type":"image\/jpeg","type":"image","subtype":"jpeg","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1261,"height":590,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output-213x200.jpg","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output.jpg","medium-width":464,"medium-height":217,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output.jpg","medium_large-width":768,"medium_large-height":359,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output.jpg","large-width":1261,"large-height":590,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output.jpg","1536x1536-width":1261,"1536x1536-height":590,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output.jpg","2048x2048-width":1261,"2048x2048-height":590,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output-826x386.jpg","card_image-width":826,"card_image-height":386,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/Regressionmodel-final-output.jpg","wide_image-width":1261,"wide_image-height":590}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

                                                              And there you have it,<\/span> you have successfully created a regression model and predicted variables. <\/span>It is<\/span> important to remember that while we all strive for perfection our data will fight us along the way and there are ways to overcome that<\/span>. Hopefully, this blog has provided you with some <\/span>new tools to help you<\/span> make decisions towards creating the best possible regression model in ArcGIS <\/span><\/span>Insights!<\/span>\u00a0<\/span><\/span>\u00a0<\/span><\/p>\n"}],"authors":[{"ID":294792,"user_firstname":"Kate","user_lastname":"Maclachlan","nickname":"Kate Maclachlan","user_nicename":"k-maclachlan","display_name":"Kate MacLachlan","user_email":"k.maclachlan@esri.com","user_url":"","user_registered":"2021-12-03 18:37:13","user_description":"Kate is a product engineer on the ArcGIS Insights team with a background in geography and GIS. In the summer she can be found sailing the Ottawa river or hiking with her dog Reggie.","user_avatar":""}],"related_articles":[{"ID":1096291,"post_author":"6771","post_date":"2020-12-23 09:22:30","post_date_gmt":"2020-12-23 17:22:30","post_content":"","post_title":"Create a COVID-19 relative risk surface","post_excerpt":"","post_status":"publish","comment_status":"open","ping_status":"closed","post_password":"","post_name":"create-relative-risk-surface","to_ping":"","pinged":"","post_modified":"2022-11-25 10:53:58","post_modified_gmt":"2022-11-25 18:53:58","post_content_filtered":"","post_parent":0,"guid":"https:\/\/www.esri.com\/arcgis-blog\/?post_type=blog&p=1096291","menu_order":0,"post_type":"blog","post_mime_type":"","comment_count":"0","filter":"raw"},{"ID":826531,"post_author":"6771","post_date":"2020-04-22 10:37:51","post_date_gmt":"2020-04-22 17:37:51","post_content":"","post_title":"Use the COVID-19 Hospital Impact Model for Epidemics (CHIME) application in ArcGIS Insights","post_excerpt":"","post_status":"publish","comment_status":"open","ping_status":"closed","post_password":"","post_name":"use-chime-arcgis-insights","to_ping":"","pinged":"","post_modified":"2022-11-25 09:47:02","post_modified_gmt":"2022-11-25 17:47:02","post_content_filtered":"","post_parent":0,"guid":"https:\/\/www.esri.com\/arcgis-blog\/?post_type=blog&p=826531","menu_order":0,"post_type":"blog","post_mime_type":"","comment_count":"0","filter":"raw"},{"ID":76121,"post_author":"6771","post_date":"2021-10-25 12:00:42","post_date_gmt":"2021-10-25 19:00:42","post_content":"","post_title":"Share your Insights analysis using ArcGIS StoryMaps","post_excerpt":"","post_status":"publish","comment_status":"open","ping_status":"closed","post_password":"","post_name":"insights-tip-share-your-analysis-using-story-maps","to_ping":"","pinged":"","post_modified":"2023-04-12 08:10:53","post_modified_gmt":"2023-04-12 15:10:53","post_content_filtered":"","post_parent":0,"guid":"http:\/\/www.esri.com\/arcgis-blog\/products\/product\/uncategorized\/insights-tip-share-your-analysis-using-story-maps\/","menu_order":0,"post_type":"blog","post_mime_type":"","comment_count":"0","filter":"raw"},{"ID":1276772,"post_author":"6771","post_date":"2021-07-26 11:03:17","post_date_gmt":"2021-07-26 18:03:17","post_content":"","post_title":"Create your first script in the Insights scripting environment","post_excerpt":"","post_status":"publish","comment_status":"open","ping_status":"closed","post_password":"","post_name":"create-your-first-script-in-insights","to_ping":"","pinged":"","post_modified":"2026-01-07 12:20:55","post_modified_gmt":"2026-01-07 20:20:55","post_content_filtered":"","post_parent":0,"guid":"https:\/\/www.esri.com\/arcgis-blog\/?post_type=blog&p=1276772","menu_order":0,"post_type":"blog","post_mime_type":"","comment_count":"0","filter":"raw"},{"ID":324122,"post_author":"6671","post_date":"2018-09-28 00:00:22","post_date_gmt":"2018-09-28 07:00:22","post_content":"","post_title":"Exploring Ocean Sensors with Insights for ArcGIS","post_excerpt":"","post_status":"publish","comment_status":"closed","ping_status":"closed","post_password":"","post_name":"use-insights-to-explore-sensor-data","to_ping":"","pinged":"","post_modified":"2019-01-17 09:59:07","post_modified_gmt":"2019-01-17 17:59:07","post_content_filtered":"","post_parent":0,"guid":"http:\/\/www.esri.com\/arcgis-blog\/?post_type=blog&p=324122","menu_order":0,"post_type":"blog","post_mime_type":"","comment_count":"0","filter":"raw"}],"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/RegressionBlog_Card.jpg","wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/RegressionBlog_Banner.jpg"},"yoast_head":"\nRegression analysis: when the data doesn\u2019t conform<\/title>\n<meta name=\"description\" content=\"A guided analysis using ArcGIS Insights to explore variables, create and evaluate regression models, and predict variables.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Regression analysis: when the data doesn\u2019t conform\" \/>\n<meta property=\"og:description\" content=\"A guided analysis using ArcGIS Insights to explore variables, create and evaluate regression models, and predict variables.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\" \/>\n<meta property=\"og:site_name\" content=\"ArcGIS Blog\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/esrigis\/\" \/>\n<meta property=\"article:modified_time\" content=\"2022-04-20T16:11:16+00:00\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:site\" content=\"@ESRI\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":[\"Article\",\"BlogPosting\"],\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\"},\"author\":{\"name\":\"Kate MacLachlan\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/50f0e3508a5e368f9c42262570d0a466\"},\"headline\":\"Regression analysis: when the data doesn\u2019t conform\",\"datePublished\":\"2022-04-06T18:24:16+00:00\",\"dateModified\":\"2022-04-20T16:11:16+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\"},\"wordCount\":8,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#organization\"},\"keywords\":[\"ArcGIS Insights\",\"regression analysis\"],\"articleSection\":[\"Analytics\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\",\"name\":\"Regression analysis: when the data doesn\u2019t conform\",\"isPartOf\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#website\"},\"datePublished\":\"2022-04-06T18:24:16+00:00\",\"dateModified\":\"2022-04-20T16:11:16+00:00\",\"description\":\"A guided analysis using ArcGIS Insights to explore variables, create and evaluate regression models, and predict variables.\",\"breadcrumb\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.esri.com\/arcgis-blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Regression analysis: when the data doesn\u2019t conform\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#website\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/\",\"name\":\"ArcGIS Blog\",\"description\":\"Get insider info from Esri product teams\",\"publisher\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.esri.com\/arcgis-blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#organization\",\"name\":\"Esri\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/04\/Esri.png\",\"contentUrl\":\"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/04\/Esri.png\",\"width\":400,\"height\":400,\"caption\":\"Esri\"},\"image\":{\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/esrigis\/\",\"https:\/\/x.com\/ESRI\",\"https:\/\/www.linkedin.com\/company\/5311\/\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/50f0e3508a5e368f9c42262570d0a466\",\"name\":\"Kate MacLachlan\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/01\/KateMacLachlan_HeadShot_UC-465x465.jpg\",\"contentUrl\":\"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/01\/KateMacLachlan_HeadShot_UC-465x465.jpg\",\"caption\":\"Kate MacLachlan\"},\"description\":\"Kate is a product engineer on the ArcGIS Insights team with a background in geography and GIS. In the summer she can be found sailing the Ottawa river or hiking with her dog Reggie.\",\"sameAs\":[\"linkedin.com\/in\/kate-maclachlan\"],\"jobTitle\":\"product engineer\",\"url\":\"\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Regression analysis: when the data doesn\u2019t conform","description":"A guided analysis using ArcGIS Insights to explore variables, create and evaluate regression models, and predict variables.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform","og_locale":"en_US","og_type":"article","og_title":"Regression analysis: when the data doesn\u2019t conform","og_description":"A guided analysis using ArcGIS Insights to explore variables, create and evaluate regression models, and predict variables.","og_url":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform","og_site_name":"ArcGIS Blog","article_publisher":"https:\/\/www.facebook.com\/esrigis\/","article_modified_time":"2022-04-20T16:11:16+00:00","twitter_card":"summary_large_image","twitter_site":"@ESRI","schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":["Article","BlogPosting"],"@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform#article","isPartOf":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform"},"author":{"name":"Kate MacLachlan","@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/50f0e3508a5e368f9c42262570d0a466"},"headline":"Regression analysis: when the data doesn\u2019t conform","datePublished":"2022-04-06T18:24:16+00:00","dateModified":"2022-04-20T16:11:16+00:00","mainEntityOfPage":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform"},"wordCount":8,"commentCount":0,"publisher":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/#organization"},"keywords":["ArcGIS Insights","regression analysis"],"articleSection":["Analytics"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform","url":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform","name":"Regression analysis: when the data doesn\u2019t conform","isPartOf":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/#website"},"datePublished":"2022-04-06T18:24:16+00:00","dateModified":"2022-04-20T16:11:16+00:00","description":"A guided analysis using ArcGIS Insights to explore variables, create and evaluate regression models, and predict variables.","breadcrumb":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.esri.com\/arcgis-blog\/products\/insights\/analytics\/regression-analysis-when-the-data-doesnt-conform#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.esri.com\/arcgis-blog\/"},{"@type":"ListItem","position":2,"name":"Regression analysis: when the data doesn\u2019t conform"}]},{"@type":"WebSite","@id":"https:\/\/www.esri.com\/arcgis-blog\/#website","url":"https:\/\/www.esri.com\/arcgis-blog\/","name":"ArcGIS Blog","description":"Get insider info from Esri product teams","publisher":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.esri.com\/arcgis-blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.esri.com\/arcgis-blog\/#organization","name":"Esri","url":"https:\/\/www.esri.com\/arcgis-blog\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/logo\/image\/","url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/04\/Esri.png","contentUrl":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/04\/Esri.png","width":400,"height":400,"caption":"Esri"},"image":{"@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/esrigis\/","https:\/\/x.com\/ESRI","https:\/\/www.linkedin.com\/company\/5311\/"]},{"@type":"Person","@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/50f0e3508a5e368f9c42262570d0a466","name":"Kate MacLachlan","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.esri.com\/arcgis-blog\/#\/schema\/person\/image\/","url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/01\/KateMacLachlan_HeadShot_UC-465x465.jpg","contentUrl":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/01\/KateMacLachlan_HeadShot_UC-465x465.jpg","caption":"Kate MacLachlan"},"description":"Kate is a product engineer on the ArcGIS Insights team with a background in geography and GIS. In the summer she can be found sailing the Ottawa river or hiking with her dog Reggie.","sameAs":["linkedin.com\/in\/kate-maclachlan"],"jobTitle":"product engineer","url":""}]}},"text_date":"April 6, 2022","author_name":"Kate MacLachlan","author_page":false,"custom_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2022\/04\/RegressionBlog_Banner.jpg","primary_product":"ArcGIS Insights","tag_data":[{"term_id":466322,"name":"ArcGIS Insights","slug":"arcgis-insights","term_group":0,"term_taxonomy_id":466322,"taxonomy":"post_tag","description":"","parent":0,"count":24,"filter":"raw"},{"term_id":37561,"name":"regression analysis","slug":"regression-analysis","term_group":0,"term_taxonomy_id":37561,"taxonomy":"post_tag","description":"","parent":0,"count":4,"filter":"raw"}],"category_data":[{"term_id":23341,"name":"Analytics","slug":"analytics","term_group":0,"term_taxonomy_id":23341,"taxonomy":"category","description":"","parent":0,"count":1329,"filter":"raw"}],"product_data":[{"term_id":36801,"name":"ArcGIS Insights","slug":"insights","term_group":0,"term_taxonomy_id":36801,"taxonomy":"product","description":"","parent":36591,"count":119,"filter":"raw"}],"primary_product_link":"https:\/\/www.esri.com\/arcgis-blog\/?s=#&products=insights","_links":{"self":[{"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/blog\/1507812","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/blog"}],"about":[{"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/types\/blog"}],"author":[{"embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/users\/294792"}],"replies":[{"embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/comments?post=1507812"}],"version-history":[{"count":0,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/blog\/1507812\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/media?parent=1507812"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/categories?post=1507812"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/tags?post=1507812"},{"taxonomy":"industry","embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/industry?post=1507812"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/www.esri.com\/arcgis-blog\/wp-json\/wp\/v2\/product?post=1507812"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}

    Don\u2019t have Insights in your organization? Get a\u00a0free trial<\/a>!<\/p>\n","image_reference":false,"layout":"standard","image_reference_figure":"","snippet":"","spotlight_name":"","section_title":"","position":"Right","spotlight_image":false},{"acf_fc_layout":"content","content":"