Curve-Fitting Methods and Messages

CURVE-FITTING METHODS AND THE MESSAGES THEY SEND LINEAR “HEY, I DID A REGRESSION.” QUADRATIC “I WANTED A CURVED LINE, SO I MADE ONE WITH MATH.” LOGARITHMIC “LOOK, IT’S TAPERING OFF!” EXPONENTIAL “LOOK, IT’S GROWING UNCONTROLLABLY!” LOESS “I’M SOPHISTICATED, NOT LIKE THOSE BUMBLING POLYNOMIAL PEOPLE.” LINEAR, NO SLOPE “I’M MAKING A SCATTER PLOT BUT I DON’T WANT TO.” LOGISTIC “I NEED TO CONNECT THESE TWO LINES, BUT MY FIRST IDEA DIDN’T HAVE ENOUGH MATH.” CONFIDENCE INTERVAL “LISTEN, SCIENCE IS HARD, BUT I’M A SERIOUS PERSON DOING MY BEST.” PIECEWISE “I HAVE A THEORY, AND THIS IS THE ONLY DATA I COULD FIND.” CONNECTING LINES “I CLICKED ‘SMOOTH LINES’ IN EXCEL.” AD-HOC FILTER “I HAD AN IDEA FOR HOW TO CLEAN UP THE DATA. WHAT DO YOU THINK?” HOUSE OF CARDS “AS YOU CAN SEE, THIS MODEL SMOOTHLY FITS THE— WAIT I DON'T EXTEND IT AAAAAA!”

A 4-by-3 grid humorously presents different curve-fitting methods as if each one were sending a message. The captions personify statistical models: some are mathematically confident, some are overfitted or improvised, and the final “house of cards” model collapses when extended beyond the available data. It is a reaction meme for jokes about data science, regression, curve fitting, and models that fail outside their training data.

• @profunctor_io / 1322 2018-09-19