Matthew Blackwell

There were some changes to R 4.2 that broke some of Zelig's examples and got it kicked off of CRAN (see #352 for details). This PR fixes those examples and...

Segfault when using `demeanlist`

1

Thanks for writing `lfe`, it's really great! One issue that I've found is that the function `demeanlist()` appears to be fairly sensitive to the way that the factors are specified,...

added tidyverse contributors

4

@xiaolong-y @liowyijia @shomiyazaki @annayrzy Can you all take a look at these names in the contributor list to see if I have the information correct? Thanks for all your help...

Model fit question in `05-prediction1` should go into `06-prediction2`.

In `01-causality1`, we check some tables, but these are not identical if a student gives the dimension of the table names. So these two should both pass: ```r table(sex =...

When there is a singularity in the first-stage of `sequential_g()`, `summary()` errors out ungracefully: ``` Error in Fhat %*% R1 %*% t(efun1) : non-conformable arguments ```

Interactions in the blip down function don't work

3

MWE: ``` r library(DirectEffects) # ACDE of the plow on female participation, with income as mediator form_main

Amelia should be able to use its output to impute a new dataset. Ideal for held out data.

When `boot.type = "none"`, `amelia()` will run the identical EM algorithm `m` times. Instead, we should probably run EM once and then call `amelia.impute()` multiple times.