« first day (2128 days earlier)      last day (2883 days later) » 

7:26 AM
@MatthewDrury Harrell's Regression Modeling Strategies and Hastie et al.'s Elements of Statistical Learning are the two books I keep on recommending.
 
 
5 hours later…
12:00 PM
@StephanKolassa: Where can one find legal free pdf of Harrell's book?
 
12:23 PM
@amoeba It seems I was mistaken - probably confused it with Hastie et al. (all those books look the same, being yellow and all). It doesn't seem to be available. Here is the book's website: biostat.mc.vanderbilt.edu/wiki/Main/RmS
 
I don't have the book by Harrell - but I observe that there are extensive course notes based on the book here - biostat.mc.vanderbilt.edu/tmp/course.pdf (part of the book's website) - which appear to be a very generous gloss of its contents (certainly enough to decide if the book will be a good investment for you)
 
 
3 hours later…
3:08 PM
How do I "paste an entire question" like @Kodiologist did above? chat.stackexchange.com/transcript/message/30292862#30292862
 
 
3 hours later…
6:12 PM
@StephanKolassa That happened automagically when I made a chat message with just the "share" permalink for the question.
It looks like the key thing is that you put the URL on a line by itself. chat.stackexchange.com/faq#formatting
 
@Kodiologist Cool, thanks!
4
Q: Strange likelihood trace from MCMC chain

C_Z_I've got a model that goes: Single parameter -> Complex likelihood function -> Log-likelihood. I executed an MCMC chain (using pymc) and plotted the trace of the parameter and the log-likelihood. The parameter estimate ended up being reasonable, but the log-likelihood plot looks strange to me. ...

(Hurray)
This question looks like it might be useful. I'm just kind of wondering whether we could find a better (better searchable) title than "strange likelihood trace".
(But I couldn't think of one.)
Anyone want to edit the title?
 

« first day (2128 days earlier)      last day (2883 days later) »