Stata Tutorial: Basic Panel Regression

[@anjalitandon4047]

very useful, the clarity in your discussions makes you a cut above the rest, Mike.

[@yoursoothingmusic1754]

this video is severely underrated. Thank you! your explanations are really clear!

[@abdulhamittayfur1076]

Thank you for the practice! very useful and realistic approach, easy to follow and understand!

[@alexanderp2375]

Hello Mr.Jonas, Right now, I'm doing a study (from 1995 quarter 1 -> 2020 quarter 4) on debt across all EU countries (panel probit models) and what I need is to read my data as panel data quarterly! I have a column on excel called Year, which takes values from 1995Q1 until 2020Q4. I used what others suggest in their videos: "generate time = tq(1995q1) +_n-1 ; format time %tq ;tsset time " , then my analysis goes from 1995q1 to 2706 q1 and doesn't stop at each country's 2020 q4. I found on another video, somebody suggesting I use: egen c_id=group(EUcountry) xtset c_id quarter, quarterly, but it doesn't work. How do you suggest I import my quarterly time series for panel data? Best regards, Alexander P.

[@andreab2114]

Very helpful, thank you! I was wondering if I can run a panel regression if my dependent variable is an index that was standardized within each group, such that the within group mean is 0 and the standard deviation is 1.

[@mikejonaseconometrics1886]

The problem I see is in removing the variation across groups, forcing their means to be equal to each other at zero. Can you standardize the index across all groups to have mean zero? Still thinking...good question!

[@andreab2114]

@@mikejonaseconometrics1886 Thank you for your reply. I can do that, but it would result in a strongly skewed distribution of the dependent variable. However, in this specific analysis, I am not interest in variation across groups (i.e. I don't want to compare countries) but only within groups.

[@user-ne7cf4gu7u]

Hello Mike! Thank you for your videos! They are very helpful. I have a problem with eststo command in Stata 14.2, it is not regognized. May you please suggest, which command can I use in 14.2 version to store coefficients and perform Hausman test afterwards. Thank you in advance!

[@mikejonaseconometrics1886]

Thanks! Try “estimates store” followed by the name you apply to the coefficient set. For example “estimates store old” following “reg y x1 x2”. Let me know if that works for you.

[@nathaliebak3706]

Hi Mike, I have got a question. I have panel data with an industry dummy variable. This variable is ofcourse time-invariant. What can I do with this variable to ensure that it will not eliminate when I do a fixed effect regression?

[@mikejonaseconometrics1886]

The Hausman-Taylor model might be a good choice in this case: kzbin.info/www/bejne/iKGXeZt6fsiAkNE

[@roseanipereiraparente3757]

Hi Do you have a tutorial on lags?