Poisson responses, log link, and linear predictor
eta_i = alpha + z_{i,1} f(x_1) + ... + z_{i,n} f(x_n)
where f(.) is a smooth function (the SAME function each time),
and the data are
x_i observations of an explanatory variable x
z_{i,j} indicators (equal to 0 or 1), KNOWN
In effect the linear predictor is a sum of a variable number of terms;
summed over a subset of the data (the subset is known and depends on i)
each summand being an evaluation of the same smooth function f
applied to the observation x_j. The function f is to be estimated.
I'd be grateful for any hints...
thanks
Adrian Baddeley
<http://maths.uwa.edu.au/~adrian/>
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news