I've only read through the chapter on transformations but this seems likely a truly excellent resource. Gratis for personal use.
The description of fitting coxph to time dependent data here seems to be better than any I've found yet.
Awesome project created by an undergrad at Brown!
An old post by Andrew Gelman on the whole "daughters make you a lefty" phenomena and how there is a potential bias introduced by the fact that having children is made of intermediate outcomes. Something that might be addressed with multilevel hierarchical models (no surprise coming from Gelman, I suppose).
Cool example of the effects of outliers on regression models.
