This web page is for error corrections and other changes to the slides and homeworks.
- (posted Nov 30).
The notes about R function
glmbb
and graphical models made no sense when applied to the horseshoe crab data, which had a quantitative predictor so didn't fit into the graphical models theory that had been presented. So we have switched to a different data set and also added a whole new section about drawing the graph for a formula and interpreting the graphs drawn. - (posted Nov 17). The code for parametric bootstrap confidence intervals was wrong in that it did not have the minus signs it was supposed to have (equation (4.1) in those notes). Now fixed.
- (posted Oct 7). Equation (6.10) in the notes on likelihood inference had a θ0 where it should have had just θ. Now fixed.
- (posted Oct 6). In the notes on Simple Chi-Square Tests for Contingency Tables a 2 was missing in the likelihood ratio test statistic in two places, in equation (2.3) and in the computation in Section 2.2.2.3. Both now fixed.
- (posted Oct 5). The calculation of the test statistic in the Wilks test for the Poisson distribution was missing a square root. Now fixed.
- (posted Oct 4). The notes about simple chi-square tests for contingency tables have had several new sections added.
- (posted Sep 23). The log likelihood function for the Poisson distribution had the fixup for the case where the first calculation gives not a number was wrong (as noticed in class). Now fixed.