I would like to implement in R a method that was proposed in a paper recently
published in Am J Epidemiol: Adjustment for Missing Confounders in Studies
Based on Observational Databases: 2-Stage Calibration Combining Propensity
Scores From Primary and Validation Data. Am J Epidemiol. 2014;180(3):308-317.
http://aje.oxfordjournals.org/content/180/3/308
My question is: Is it possible to obtain the U_i's in the lines below (from the
web-appendix - somewhat modified) in R if the model G is a cox or poisson
regression model?
Let gamma be a parameter estimate obtained from a model G based on data for
subjects {i=1,...,N}. E.g. the effect of an exposure on an outcome.
Let < \sum{i=1}^N Q_i > be the likelihood function and < I > the negative
hessian (information) matrix of the likelihood for the model .
Let < U_i > be the efficient score of gamma for the i'th subject, defined by
the subset of components in the vector < I^{-1}Q_i > that corresponds to the
parameter gamma.
Then < {U_i, i=1,...,N} > are variables with identical and independent
distributions and < gamma = \sum_{i=1}^N U_i > asymptotically.
Best regards,
Vidar Hjellvik
[[alternative HTML version deleted]]
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
