Finding maximum likelihood estimates of parameters of a proportional hazards model.

78 Views Asked by At

Consider the proportional hazards model $h(t,x)=\lambda(t)\exp\{\beta'x\}$. Let $\Lambda(t)=\int_{-\infty}^t\lambda(u)du=\exp\{\alpha t\}$. Suppose there are $n$ uncensored and $m$ censored observations. I need a methodology for computing the maximum likelihood estimates of the parameters $\alpha$ and $\beta$.

I don't need complete mathematical details. Just suggest me a method. Thanks in advance!

1

There are 1 best solutions below

0
On BEST ANSWER

I think you can find this usefull! http://www-personal.umich.edu/~yili/lect4notes.pdf