Hengshi Yu, Tahmeed Tureen, Weiqi Zhou
This is a C++ course project for the statistical computing course.
Both geec_oneloop.cpp
and geec_twoloop.cpp
provide the C++ implementation of generalized estimating equations (GEE) with Gaussian outcomes.
After compiling the two cpp files into programs, we can implement GEE on the example data as
./gee/geec_oneloop ./data/example_data/studyId.csv ./data/example_data/outcomes.csv ./data/example_data/designX.csv
./gee/geec_twoloop ./data/example_data/studyId.csv ./data/example_data/outcomes.csv ./data/example_data/designX.csv
Users can optionally set an additional fourth arguments of weights in GEE.
Compared with several R-based GEE packages on several simulated datasets (in /data/simulation.R
), the geeCpp gains much better time efficiency.