I am trying to fit a weighted Cox regression model in the spirit of Peter Sasieni (1993) "Maximum Weighted Partial Likelihood Estimators for the Cox Model" and D. Y. Lin (1991) "Goodness-of-Fit Analysis for the Cox Regression Model Based on a Class of Parameter Estimators".
I know PHREG allows case weights via WEIGHT statement, but I would like to accomplish is weighting the score function from weighted log-rank test (e.g. Fleming and Harrington).
I found a document written by Terry M. Therneau "A Package for Survival Analysis in S" where in Section 3.3.7, he explains that to estimate beta from a weighted cox model an offset -log(weight) has to be included. Standard error need to be also corrected.
I include link to the document Therneau: https://www.mayo.edu/research/documents/tr53pdf/doc-10027379
I did not find any other source in the web to fit this model from a programmatic point of view.
Is anybody experienced with this kind of estimation?
Thank you
