This web page is for error corrections and other changes to the slides and homeworks.
- (posted Sep 25). In the section about exact two-tailed tests for the binomial distribution the notes formerly said this makes no sense. But now they say they make no sense except when the hypothesized success probability under the null hypothesis is 1 ⁄ 2, because then the distribution of the test statistic under the null hypothesis is symmetric. Needed for homework problem 1-7 (e).
- (posted Sep 29). Not an error, just an addition. A student asked what the null and alternative hypotheses were for the hypothesis tests in the Notes on Chi-Square Tests. So they have now been specified.
- (posted Oct 2). Lots of changes made to Notes on Chi-Square Tests.
- (posted Oct 4). Graph of minus the log likelihood for Cauchy location model added to Notes on Maximum Likelihood Estimation.