(Untitled)

[augustus]

ok, here it is. It gets pretty bad towards the end, but as long as I have something to turn in tomorrow, that is ok, I have two weeks to make any corrections on it. Now I will go snuggle with my husband for a little before he has to get up and go to work.

Here is a piece of crap. Read if you want.

Comments

[sqrrlsrant]

Hmmmmm, long night was it? Smoking crack were you? How many posts is that?

[augustus]

The estimates for all 13 models are derived in SAS through a procedure called nlmixed. The estimates are the result of a maximum likelihood function, and it is the same ML function for all the models, then only thing that changes is how the parameters vary (e.i. linear, constant, first-year then constant). The results of models C and D are actually more of what I was hoping for. An estimate of .5 tells me jackshit.

My advisor isn't going to use this as a funding proposal, although a big chuck of their budget does come from the LDWFFRD because he writes a survival analysis every year for them and helps them with their research.