{"id":3253,"date":"2023-03-10T13:41:11","date_gmt":"2023-03-10T12:41:11","guid":{"rendered":"https:\/\/www.microdata.no\/?p=3253"},"modified":"2023-11-22T14:34:48","modified_gmt":"2023-11-22T13:34:48","slug":"new-analysis-functionalities-diff-in-diff-and-pooled-panel-regression","status":"publish","type":"post","link":"https:\/\/www.microdata.no\/en\/new-analysis-functionalities-diff-in-diff-and-pooled-panel-regression\/","title":{"rendered":"New analysis functionalities: Diff-in-diff and pooled panel regression"},"content":{"rendered":"\n

It is now possible to perform diff-in-diff analyzes using the new regress-panel-diff command. In addition, it is now permitted to run pooled panel regression.<\/em><\/strong><\/p>\n\n\n\n\n\n\n\n

Pooled panel regression<\/h2>\n\n\n\n

Linear panel regressions have until now been possible through the command regress-panel<\/code>, using the alternatives fixed effects (fe<\/code>), random effects (re<\/code>) and between effects (be<\/code>).<\/p>\n\n\n\n

We are now introducing the new option pooled<\/code> which makes it possible to run pooled panel regressions. This is equivalent to running ordinary linear regression on panel datasets, where the time dimension is disregarded (considering all observations as separate units).<\/p>\n\n\n\n

Example:<\/p>\n\n\n\n

\"\"
Random selection with panel extraction for the years 2018-2021. Dependent variable = annual salary, using marital status = married, place of residence = Oslo, and educational level equal to a master’s degree or higher as explanatory variables. Also years are included to control for time.<\/figcaption><\/figure>\n\n\n\n

Diff-in-diff<\/h2>\n\n\n\n

Diff-in-diff is a widely used form of analysis where the effect of a “treatment” is analyzed by comparing the change in the average value for a continuous\/rankable response variable before\/after the time of treatment. This is done for two groups: the treatment group and the control group. Finally, the difference between the two groups is calculated.<\/p>\n\n\n\n

The following preparatory steps must be followed before running a diff-in-diff analysis:<\/p>\n\n\n\n

    \n
  1. Create a panel dataset through the command import-panel<\/code> or by converting from “wide” format to “long” format through the command reshape-to-long<\/code><\/li>\n\n\n\n
  2. Create a group variable with the value 1 for the treatment group and 0 for the control group<\/li>\n\n\n\n
  3. Create a treatment variable that is set to 0 for all times before the treatment time, and 1 for all times starting from the treatment time<\/li>\n<\/ol>\n\n\n\n

    After following steps 1. – 3. the command regress-panel-diff<\/code> is used.<\/p>\n\n\n\n

    The dependent variable is listed first. It must be continuous or rankable. The group and treatment variables need to be listed as numbers 2 and 3.<\/strong> This is a prerequisite for the analysis to be carried out correctly. Other independent variables are listed at the end (optional).<\/p>\n\n\n\n

    The result from regress-panel-diff<\/code> shows a standard panel regression table with model measures and coefficient values. The diff-in-diff value (so-called ATET value \u2013 average treatment effect of the treated<\/em>) corresponds to the coefficient value of the interaction term for the two dummy variables which indicate respectively group and treatment.<\/p>\n\n\n\n

    Example:<\/p>\n\n\n\n

    \"\"
    Random selection with panel extraction for the years 2018-2021. Uses women as the treatment group and men as the control group. The treatment time is set to 2020. Diff-in-diff value (ATET) is equal to -1991.77 when controlling for marital status = married, place of residence = Oslo, and educational level equal to a master’s degree or higher. The ATET value is not significant in this case. <\/figcaption><\/figure>\n\n\n\n

    The command regress-panel-diff<\/code> is equivalent to running regress-panel<\/code> with the option pooled<\/code> where the group and treatment variables are included as interaction terms as well as separate dummies (use the characters ## to express this).<\/p>\n\n\n\n

    Example:
    regress-panel-diff salary group treatment married oslo high_edu<\/code>
    gives the same result as
    regress-panel salary group##treatment married oslo high_edu, pooled<\/code><\/p>\n\n\n\n

    The following options are available for regress-panel-diff<\/code>:<\/p>\n\n\n\n