8
votes
Accepted
What are some important papers using difference-in-difference in economics?
This is far from exhaustive, but hopefully it can get you started! (I'd recommend skimming most of them, then really focusing on those where the application and/or modeling environment is uniquely ...
- 1,380
6
votes
Accepted
What shoud we do when the expected treatment overlaps control sample in DID?
Your first quote seems to be given as an explanation of how they constructed Figure 1 (however I cannot be sure since you did not state pagenumber for the quote). Anyway, Figure 1 compares a treatment ...
- 3,788
5
votes
Accepted
In difference-in-differences approach, would it be possible to add a dummy variable indicating whether a data point belong to treatment group?
It depends on the estimated model. Let me use the example provided in Mostly Harmless Econmetrics.
The experiment
Suppose we are interested in the effect of the minimum wage on employment. Card and ...
- 6,990
5
votes
Accepted
Is program evaluation (DiD, RD) a structural estimation?
I skimmed through the slides you linked.
I think professor Haile in these slides is trying to introduce the concepts of "structural" and "reduced-form" models in a very broad sense....
- 131
4
votes
How to correct selection bias in an econometric analysis?
The parallel trends assumption of Diff-in-diff is essentially an assumption that there is no selection bias. If you are curious about the parallel trends assumption, I encourage you to ask about it in ...
- 3,768
3
votes
Accepted
Reverse DiD: or using always treated as control
In principle yes, but you would not use 'reverse treatment' but standard terminology. Here just withdrawal of some stimuli is the treatment. That is you would still code the withdrawal as $D=1$.
For ...
- 57.7k
3
votes
Accepted
Difference in differences with one post and two treatment variables
Let $x_{it}$ denote after rejection, $z_{it}$ denote after granted, $y_{it}$ the outcome, and $w_i$ is US origin (which should not change over time and doesn't have a $t$ subscript).
The two-way fixed ...
- 3,768
3
votes
Accepted
What should we do if we find the anticipation effect?
The no anticipation assumption requires an individual outcome to not depend on future treatments.
The best way you can do in the presence of anticipation effects is to use an disaggregated approach ...
- 210
3
votes
Two-year difference-in-differences model design
Is there some literature about two-year difference-in-difference (DiD) models?
Of course.
The traditional two-group/two-period DiD design was quite popular 25 years ago, though it's less common today ...
- 459
3
votes
Accepted
What does "inter alia" in "difference-indifference" testing?
It means "among other things". Its not an economics term, its standard English. It is somewhat pretentious, having been borrowed from Latin.
Can I suggest that next time you don't understand ...
- 219
2
votes
Are there any papers which used the event study methodology NOT in a financial market context?
See this Github page from David Novgorodsky and Bradley Setzler (University of Chicago) and the companion.
- 61
2
votes
Accepted
Should we test for an anticipation effect in difference-in-differences when a natural event occurs?
A natural event is merely 'mimicking' the features of random assignment. As in your example, a tsunami is a natural event and may impact some regions, but not others. Is anticipation not a concern in ...
- 459
2
votes
Accepted
Diff-in-diff parallel trends with a positive outcome
The purpose of the parallel trends assumption is that it allows you to use changes in the control group to estimate what changes in the treatment group would have been in the counterfactual case where ...
- 636
2
votes
How to define treatment & control groups properly?
It looks to me like your control group would be all families with 1 or 0 kids. This is not ideal as you mention.
If you observe all households before and after the transfer, then you could do a diff-...
- 3,768
2
votes
Diff in Diff: Why do we control for unit and time fixed effects instead of just treatment status and pre/post indicator?
The 2-period model you wrote and the fixed effects (FE) models are equivalent for 2-periods and one treatment. The advantage of fixed effects, is that it extends to multiple periods and treatments (e....
- 6,246
2
votes
Accepted
How to correct selection bias in an econometric analysis?
As the other answer mentioned, if the Diff-in-diff assumptions hold, it should not be a problem.
However, if you think selection might be invalidating your diff-in-diff assumptions, you have two ...
- 6,246
2
votes
Accepted
Difference between unconfoundedness and parallel trends
You're pretty much correct - the two assumptions are very similar.
The only difference is that the Parallel Trends assumption allows for differences between the treatment group and the control group (...
- 402
2
votes
Accepted
Diff-in-diff with different treatment years, what would be time dummy for control?
Assume you have a dataset $\{Y_{it},X_{it},D_{it}\}$ where $Y_{it}$ is the dependent variable, $X_{it}$ are control variables and $D_{it}$ is an indicator which for observational unit $i$ is equal to $...
- 3,788
1
vote
Pooled difference-in-differences with limited panel
Yes, it does. The individual effects will be imprecisely estimated, but those coefficients are not of primary interest.
The assumption you need is strict exogeneity, $E[\epsilon_{i,t}|X_i]=0$, where $...
- 3,768
1
vote
Is matching combined with Diff-in-Diff a bad idea?
Is matching combined with Diff-in-Diff a bad idea?
No.
But, as with anything, it depends on the context. One point to consider is that you should match on pre-treatment characteristics, because ...
- 6,246
1
vote
Accepted
Is Difference-in-Differences still valid when treat or control is determined by subjects?
It is definitely an issue that needs to be considered.
So, you are right that if the parallel trends assumption is not violated the approach is fine. Note the assumption in other words is that "...
- 6,246
1
vote
Diff-in-diff with two control groups: compare parallel trends
Background
The typical thing to do is visually inspect the pre-treatment trends for the control and treatment groups. Whether you want it or not, you might be biased when looking at the visual ...
- 939
1
vote
Diff-in-diff with two control groups: compare parallel trends
I'm not understanding-- could you please talk a little about why both can't be used here? It's been a while since I've really worked with DD, but as far as I'm aware, there's not real metric you can ...
- 129
1
vote
Discussing Difference-in-Difference Assumption of the treatment assigment
The first source you found is correct, parallel trend is not sufficient, you can find the same assumption mentioned in multiple places (e.g here). One of the identifying assumptions of DiD is that:
$$...
- 57.7k
1
vote
Heterogeneity of time invariant characteristic in event study model
If you have a time-constant covariate $X$, you could include it but drop the individual fixed effects $\sigma_i$. (There could be perfect multicollinearity if you kept them.) Since you mentioned that $...
- 2,385
1
vote
Why do we need Exogeneity of Treatment in a Diff in Diff
There is nothing in the econometrics that prevent you from measuring the difference in the difference when the treatment is endogenous. The concern is about interpreting the difference in the ...
- 16.3k
1
vote
Accepted
Differences in Differences with Small Violations of Parallel Trends
If you have violations of the parallel trend assumption then you will have biased estimates of the causal effect (assuming you satisfy the other assumptions). If you know the maximum that the two ...
- 16.3k
1
vote
Difference-in-difference robust to heterogeneous treatment effect - Gendron-Carrier et al. specification
My advice is to do a stacked diff-in-diff. For each treated unit, randomly assign a control unit. This forms a group, $g$. There will be several such groups.
$$y_{igt} = \sum_{s}\beta_sT_{ig}d_{gs}+\...
- 3,768
1
vote
Accepted
Do anticipation effects almost always bias the DiD ATE or ATT towards a null result and increase standard errors?
[a synthesis of Evangelos, Michael and my comments / posts]
You are right if you have a simple, two-time period diff-in-diff and you believe parallel trends is true OR if you have staggered/...
- 270
1
vote
How to explain the Difference-in-Difference for two groups two periods to undergraduate students?
If we compare two states (like two of the 50 states I mean) to see if some big policy change improves something, would that work? Where one state has the policy and one doesnt?
TX has the policy and ...
- 121
Only top scored, non community-wiki answers of a minimum length are eligible