Greg and Patrick explore the generalized linear model as a powerful framework for building regression models for binary and other discretely distributed dependent variables. Along the way they also discuss stealing property, statistical conspiracy theories, mic drops, coming uncorked, getting punched by biostatisticians, big logistic, tapping out, the Oakland Raiders, being 8.5 feet tall, sheep […]
S3E17: Logistic Regression: 2 Logit 2 Quit Read More »
In today’s episode, Patrick and Greg use the context of COVID rapid tests to discuss issues of sensitivity, specificity, positive and negative predicted values, and the generally questionable utility of test accuracy information. Along the way they also discuss escape rooms, C4, Embassy Suites, palak paneer, 93% accurate, astragali, SAT prep courses, the volume of
S3E16: Your COVID Rapid Test Result: Are You Positive You’re Positive? Read More »
In today’s episode Greg & Patrick discuss the causes, consequences, and potential solutions associated with negative residual variances in factor analyses, a condition commonly called a Heywood case. Along the we way they also discuss vegetarian pepperoni, Jaws Part 2, coffin seat belts, balancing a ship, bad puns, sterilizing needles, dead canaries, hitchhikers, legal depositions,
S3E15: Heywood You Help Me With Negative Residual Variances? Read More »
