Embedded Files
This page consists of the my research focused functions that are also stored in the Github. This page is not complete yet.
Covariate-wise optimal cutoff and diagnostic accuracy measures. This contains functions to estimate covariate-specific AUC and common optimal threshold estimates. The manuscript associated with the functions is:
Ghosal, S., 2025. Impact of methodological assumptions and covariates on the cutoff estimation in ROC analysis. Biometrical Journal, 67(3), p.e70053, https://doi.org/10.1002/bimj.70053
Concave PV-based ROC. This repository is incomplete. Once complete, this will contain function to impose concavity on ROC curves.
Screening for multicollinearity. This is not something novel however practical for data analysis. This repository contains functions to check for multicollinearity in the dataset that consists of both numerical and categorical data.
Bayesian hierarchical spline-based hurdle model. This repository is also in the process of being made. This will contain codes to implement a spline-based Bayesian hierarchical hurdle model to account for multi-level spatiotemporal variability of zero-inflated count data. The paper associated with the code is:
Ghosal, S., Lau, T.S., Gaskins, J. and Kong, M., 2020. A hierarchical mixed effect hurdle model for spatiotemporal count data and its application to identifying factors impacting health professional shortages. Journal of the Royal Statistical Society: Series C (Applied Statistics), 69(5), pp.1121-1144, https://doi.org/10.1111/rssc.12434
Page updated
Google Sites
Report abuse