I have:
- $\boldsymbol{r}$, a $(1 \times N)$ vector of times
- $\gamma$, a $(p \times 1)$ vector of regression parameters
- $V$, a $(N \times p)$ matrix of covariates
I want to get the expression to estimate the $\gamma$ parameters. For that I have to solve the following score equation $$\frac{\partial}{\partial\gamma}\left[V\gamma -\boldsymbol{r}\exp(V\gamma)\right]=0$$.
How to do that? I’m not sure how to differentiate the exponential of the matrix times the vector of parameters.