report the casual effect of each of leaves in the plot

[zhong1969]

I have a question related to the codes of “grf”.

I used the following codes to plot a tree:

tree3.plot = plot(get_tree(forest3.iv, 1))
cat(DiagrammeRsvg::export_svg(tree3.plot), file='plot_non_cluster.svg')

In each of leaves in the plot, it reported size, avg_Y, avg_W and avg_Z. Could you please kindly tell me how to report the casual effect of each of leaves in the plot?

Great thanks!

[erikcs]

There is currently no built in functions to report that, but this thread may be useful: #238 . The are also some resources in the documentation: https://grf-labs.github.io/grf/articles/diagnostics.html#assessing-fit

[erikcs]

It will eventually be possible to do this with #281. Note that you can also use the package policytree to fit a shallow tree on the double robust scores, see the package documentation here

[fateme74]

Hi,
when we run these lines of codes below,is str.error the MSE?
If not how we can access MSE or MAE?
#>
#> Best linear fit using forest predictions (on held-out data)
#> as well as the mean forest prediction as regressors, along
#> with one-sided heteroskedasticity-robust (HC3) SEs:
#>
#> Estimate Std. Error t value Pr(>t)
#> mean.forest.prediction 1.01173 0.12993 7.7867 5.48e-15 ***
#> differential.forest.prediction 1.19326 0.10122 11.7885 < 2.2e-16 ***
#> ---
#> Signif. codes: 0 '' 0.001 '' 0.01 '' 0.05 '.' 0.1 ' ' 1

[fateme74]

Hi,
after fit a causal forest how we can access to final leaf and save observation that fall in final leaf to make other prediction on them?(Actually I want to fit a causal forest on my data to find subgroups and I need to do other prediction on each subgroup)