It is a slow week, and a week of Thanks, so let's spend our time with some fun things. Like a fascinating study by Greg Ellifritz on "An Alternate Look…
Review! We left off with comparing the standard, out-of-the-box linear regression with our multinomial predictive observable model. The great weaknesses of the regression were probability leakage (giving positive probability to…
Review! Last time we created four models of CGPA. Which is correct? They all are. Why? I should ask as a homework question, but I'll remind us here. Since all…
Review! Last time we successfully computed our model (8): (8) Pr(CGPA = i | guesses of new measures, grading rules, old obs, model), where i = 0,...,5. The model…
Kevin Gray is back with another question, this time about priors. His last led to the post "Was Fisher Wrong? Whether Or Not Statistical Models Are Needed." (The answer was…
This is mainly a catch up post for new readers who won't recall the old days. Listen to the podcast at YouTube, BitChute, or Gab. I'm late on this! STATE…
This is not a big story, but it's worth highlighting because of the pernicious effects of science journalism, which suffers from the same faults as political journalism. Emphasis on extremes,…
"Masks don't work" is a proposition, like all propositions, that carries implicit definitions. In this case it means mask mandates don't reduce the spread of respiratory diseases. It does not…