Timeline for Difference in Difference - Does the control units need to be similiar?
Current License: CC BY-SA 3.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 5, 2017 at 12:54 | comment | added | Andy | Ok, now I understand. Yes. In this case the log transformed variable will provide linear paths for the treatment and control groups and the difference-in-differences assumption is satisfied. | |
| Jan 4, 2017 at 20:18 | comment | added | B_Miner | Meaning if two markets are growing at 20% each period....cant we simply take the log of the response variables and run a normal D-i-D? | |
| Jan 4, 2017 at 20:05 | comment | added | B_Miner | I was thinking actually even simpler, if there was just a transformation of the response variable to then allow the normal DiD to work? | |
| Jan 4, 2017 at 7:33 | comment | added | Andy | According to the linked paper, yes. They establish conditions for when parallel growth is sufficient for their difference in differences estimator to work. I have not yet used their method though and it's a very recent paper so not everyone may like it | |
| Jan 4, 2017 at 1:42 | comment | added | B_Miner | Andy, if we dont have access to control markets / states / whatever that have parallel trends, but they are growing at similar rates can a DiD still be used, but we need to log the data (natural log)? | |
| Dec 26, 2016 at 14:51 | vote | accept | B_Miner | ||
| Dec 23, 2016 at 15:59 | comment | added | Andy | Actually, I never heard of it but perhaps it's just because I'm too stuck in my field :-) could also be that I know it under a different name. Personally I use DiD a lot in my research because it's a tractable method that can be easily understood by policy makers and general audiences. | |
| Dec 23, 2016 at 15:12 | comment | added | B_Miner | Andy - are you familiar with Interrupted Time Series analysis? I took this course : edx.org/course/… and it was stated that there are no parallel trends assumptions since it is being modeled explicitly. Curious if you have used this technique as a superior one to d-i-d? | |
| Dec 23, 2016 at 15:05 | vote | accept | B_Miner | ||
| Dec 26, 2016 at 14:51 | |||||
| Dec 23, 2016 at 11:08 | history | edited | Andy | CC BY-SA 3.0 |
added 22 characters in body
|
| Dec 23, 2016 at 5:52 | comment | added | Andy | That's right. This is what the linked paper is about though: developing a class of DiD estimators that can exploit parallel growth rather than parallel paths/trends. | |
| Dec 23, 2016 at 1:08 | comment | added | B_Miner | Hi Andy. So, when we say the D-I-D assumption is parallel trends, we are talking about the same slope (rise/run) and if two markets (or states or whatever) instead have the same growth rate (percentage change) these would not be candidates for D-i-D? | |
| Dec 22, 2016 at 18:59 | history | answered | Andy | CC BY-SA 3.0 |