You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If we have a discrete treatment and we use bootstrap, then it's possible that some subsamples will not have the full set of treatments, in which case some of our code will break. This becomes more likely as the number of replicates increases or the number of samples (or number of samples with one particular treatment) decreases.
Instead, we need to implement a variant of bootstrap that stratifies by treatment, so that we are sure to keep some examples with each treatment in each bootstrap sample.
The text was updated successfully, but these errors were encountered:
If we have a discrete treatment and we use bootstrap, then it's possible that some subsamples will not have the full set of treatments, in which case some of our code will break. This becomes more likely as the number of replicates increases or the number of samples (or number of samples with one particular treatment) decreases.
Instead, we need to implement a variant of bootstrap that stratifies by treatment, so that we are sure to keep some examples with each treatment in each bootstrap sample.
The text was updated successfully, but these errors were encountered: