New answers tagged

0

DID estimation uses four data points to deduce the impact of shock on the treated population: the effect of the treatment on the treated. OTOH, the required number of sample points does depend on objects. If you are doing exploratory analysis just to see if one model (say linear in a covariate) looks better than another (say a quadratic function of the ...


0

In repeated measures designs, the subjects are their own controls because the model assesses how a subject responds to all of the treatments. However, treatments order effects can interfere with the analysis’ ability to correctly estimate the effect of the treatment itself, (confounding). It seems mixed models are becoming more widespread, but there is a lot ...


2

From the estimation perspective, pweights is internally used the same way as any other weight. in OLS: $$ min_{\beta} = \sum{(y-\beta X)^2 * w} $$ The only difference with other methods is during the estimation of standard errors. When you use pweights, is like requesting robust standard errors. (sandwich formula). For further details on how exactly weights ...


0

As this question has so many views and ranks highly in google search, I would like to add to the excellent answer before (+1) that one has to be careful about actually interpreting the effects as within effects. Probably the most common empirical mistake in applied work. There is a very nice paper on this (not mine): Mummolo, J., & Peterson, E. (2018). ...


0

You can use a random-effects model in place of a paired t-test. Say you have two sets of measurements denoted by $i=1$ and $i=2$ one of the population mean $\mu_1$ and one of the population mean $\mu_2$. The measurement consists of points $y_{ij}$, which have errors within participants $\epsilon_j$ and from test to test $\epsilon_{ij}$ $$y_{ij} = \mu_i + \...


0

Another way to parameterize it would be to fit a longitudinal ANCOVA with post-test RT as the outcome, and pre-test as a time-varying predictor. If your main interest is in tracking post-test RT as a function of age, this gives you a direct measure of the age effect at post test, while controlling for the pretest. This seems more straightforward and should ...


0

Many of your questions cannot be answered definitively and require you to make some ad hoc decisions prior to estimating your model. I will address each question and comment, in turn. I am in the process of setting up a difference-in-differences analysis (DiD) to evaluate a particular policy. It was launched on the year 2000, and it basically is relevant in ...


1

I don't understand exactly why, but I solved the issue with the panel data. Apparently, the problem was that I was adding "Company" in the panel index. In other words: I was using this: pdata <- pdata.frame(mydata, index = c("Company","t")) And I should have been using this: pdata <- pdata.frame(mydata, index = "t&...


1

From my understanding of your question, this sounds like it might better be handled with a 2-event survival model. You evidently have 2 events, the first representing finishing the "mechanical part of the assembly" and the second representing full assembly. You specify the data and model in a way that requires the first event to occur before the ...


0

I am using 2016 as a reference point, so it seems to me at first glance that pooled cross-sectional data is the way to go with time dummies. Perhaps. It appears your data is more aptly described as repeated cross-sectional data. The survey repeatedly samples a new subset of individuals in each survey wave. You could certainly estimate a standard linear ...


Top 50 recent answers are included