Stop Doing Time Series

STOP DOING TIME SERIES •DATA WERE NEVER SUPPOSED TO BE PLOTTED AGAINST TIME •YEARS OF ACF PLOTS yet NO BETTER METHOD FOUND THAN LOOKING AT THE DATA WITH YOUR EYES •Wanted to predict something anyway for a laugh? We had a tool for that: it was called LINEAR REGRESSION •"My data are not iid, it has autocorrelation" - Statements dreamed up by the utterly Deranged LOOK AT WHAT THEY WANT US TO BELIEVE stationary (phi=0.8) BIC = k ln(n) − 2 ln(L̂) They say this is stationary, BUT ITS CLEARLY MOVING Back in my day BIC made LIGHTERS and PENS "Wiener process", "Dickey-Fuller test", THEY KNOW YOU'LL BELIEVE ANYTHING "Hello I would like apples please" THEY HAVE PLAYED US FOR ABSOLUTE FOOLS

A parody conspiracy-style meme attacks common time-series analysis practices, showing charts, statistical formulas, and labels claiming that supposedly stationary data are visibly moving. It mocks autocorrelation, linear regression, ACF plots, BIC, the Wiener process, and the Dickey-Fuller test as arbitrary academic inventions, ending with an absurd fake dialogue and the declaration that statisticians have fooled everyone. It is typically shared as an ironic reaction by people joking about confusing or overcomplicated statistics. EN: A parody conspiracy-style meme attacks common time-series analysis practices, showing charts, statistical formulas, and labels claiming that supposedly stationary data are visibly moving. It mocks autocorrelation, linear regression, ACF plots, BIC, the Wiener process, and the Dickey-Fuller test as arbitrary academic inventions, ending with an absurd fake dialogue and the declaration that statisticians have fooled everyone. It is typically shared as an ironic reaction by people joking about confusing or overcomplicated statistics.

• @right_wing_dev_squads / 7458 2023-08-24