Timeline for Difference-in-difference regression with pooled cross sectional data - fixed effects
Current License: CC BY-SA 4.0
7 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Jul 2, 2023 at 23:49 | history | edited | Thomas Bilach | CC BY-SA 4.0 |
Minor textual edit.
|
Jul 2, 2023 at 8:21 | history | edited | User1865345 | CC BY-SA 4.0 |
added 1 character in body
|
Mar 30, 2022 at 12:46 | comment | added | kaliscle | or do you think I should include a fixed time effect for the construction year of the dwellings? | |
Mar 29, 2022 at 18:54 | comment | added | kaliscle | I actually included the years of offer as the time dummy set which corresponds to 10 years (2011 as reference category). If adding a series for year dummies, how could I compute them? I mean based on which argument would it be 0/1 in each year 2011-2021? | |
Mar 28, 2022 at 19:41 | comment | added | Thomas Bilach | No problem. Including the time dummies seems wise. But I do have a concern. Why do you use the “date of offer” as standing in for the times fixed effects? Why not use a series of year dummies? | |
Mar 28, 2022 at 11:37 | comment | added | kaliscle | Thank you for your answer and the model specifications. I tried the different adjustments within the lm-function. If I include the zipcode + year dummies, the R² is highest compared to the model with only a location fixed effect and the p-value of the interaction term braked:rba is <0.01 whereas with only a location fixed effect the p-value is <0.05. The standard error doesn't change for the interaction term. As you pointed out, the value of the estimates just changed from -0.006 (location dummies) to -0.009 (year+location dummies) | |
Mar 19, 2022 at 4:30 | history | answered | Thomas Bilach | CC BY-SA 4.0 |