Some feedback and some questions about cinnamon HOT 3 CLOSED

On Sun, Nov 28, 2021 at 5:19 AM Le Faou ***@***.***> wrote: Hi Eric, Thanks for the feedback! From the experiments I have made the problem with the statistical test used in the context of data drift detection is that when samples are large the test generaly detect a data drift (in real life it is rare that a phenomenom is fully stationnary, at least in the cases I have been studying). About multiple comparison problem, the solutions would be the same as when we have multiple comparison in the context of A/B test (Bonferroni correction, etc.). Also note that if the test variable is discrete then Chi2 test can handle the multiclass scheme. I agree permutation test is an interesting idea, though it may be require many computations in some cases. Also, we may find some natively multi-dimensional statistical test in some situation (e.g. for the KS test : https://inis.iaea.org/search/search.aspx?orig_q=RN:18067274) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAHCZPJWXCWM7D5DQAC5W7LUOH6ZZANCNFSM5I4WGVFA> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.