Tabulating GLM summaries for each predictor, for multiple GLM fits

Asked Jul 20 '21 at 18:25

Active Jul 20 '21 at 18:25

Viewed 114 times

All, I am looking for an efficient way to do the following in R.

I have a list of R GLM objects G that all use the same model matrix X. Each GLM object just corresponds to a different response variable Y1... Yp.

I would like to create a summary list S such that S[[i]] is a data frame with the p^th row containing the summary(G[[p]])$coefficients[i,] for the p response variables in each row. Thus, the length of list S, length(S), will correspond to the total number of columns in X. The number of rows of S[[i]], nrow(S[[i]]), will equal the number of response variables.

Currently, I am running summaries for each object, and then extracting corresponding to each column in X. So I am wondering if there is a more efficient way to approach this.

asked Jul 20 '21 at 18:25

mskb

This isn't bad but `broom::tidy()` might be a good starting point – Ben Bolker Jul 20 '21 at 18:33
1

`sapply(G, coef)` ? – Ben Bolker Jul 20 '21 at 18:33
@BenBolker, that's great and so simple!! Thank you! But I see it only gives back the coefficients. The standard error, test statistic and p-value summaries are missing. – mskb Jul 20 '21 at 18:36
OK. And the number of *columns* of `S[[i]]` is the number of columns of the summary ? – Ben Bolker Jul 20 '21 at 21:22
Now I'm wondering why `lapply(G, summary)` isn't the answer to your question ... ? – Ben Bolker Jul 20 '21 at 21:36

Tabulating GLM summaries for each predictor, for multiple GLM fits

0 Answers0