If you’re building a time series model and you want any hope of statistical significance, do your analysis on non-overlapping periods.https://twitter.com/macrocephalopod/status/1460302495507525635 …
-
newey west has entered the chat
I feel like newey-west is okay if you have either natural autocorrelation of your independent variable OR deliberate autocorrelation from overlapping periods but if you have both you’re probably screwed. Idk, just a vibe though.
-
-
i feel that vibe too
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
With a largeish sample Hansen-Hodrick is an alternative I’ve found worth digging into. There’s debate that NW subsumes it, but HH procedure was designed to counteract this very problem. Try to avoid overlapping data like the plague though
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.