ehsanx Goto Github PK

python-for-epidemiologists

Tutorial in Python targeted at Epidemiologists. Will discuss the basics of analysis in Python 3

qgcomp

QGcomp (quantile g-computation): estimating the effects of exposure mixtures. Works for continuous, binary, and right-censored survival outcomes. Flexible, unconstrained, fast and guided by modern causal inference principles

rcf

heterogeneous treatment effect estimation with causal forests

rmdwriting

RMD sample files

rmrf-code

Implements code for RMRF manuscript

rsmatch

Risk set matching in R

sargc-timethods

TI Methods Speaker Series in collaboration with the Student and Recent Graduate Committee (SARGC) of the Statistical Society of Canada.

sbdecomp

scientific-writing-for-health-research

Adaptation and Deployment of OER for Communicating Scientific Research Findings in Health Sciences Education

sdrcausal

SDRcausal is a R Package that provides semiparametric estimators of Average Causal Effects, using sufficient dimension reduction for nuisance model estimation.

seasym2020

Slides for Seattle Symposium Short Course 2020

sejdemyr.github.io

Personal website

sequential_trials

R code for implementation of the simulation study described in the paper: "Causal inference in survival analysis using longitudinal observational data: Sequential trials and marginal structural models"

ser-tidyverse-2020

ser2021-workshop

Materials for the workshop "Targeted Learning in the tlverse: Causal Inference Meets Machine Learning" at the 2021 Society for Epidemiologic Research (SER) Meeting

ser2021_mediation_workshop

Materials for the workshop "Causal Mediation: Modern Methods for Path Analysis" at the 2021 Society for Epidemiologic Research Meeting

setoguchi

setoguchi simulation algorithm from 2008 paper

sim-tmle-tutorial

Targeted Maximum Likelihood Estimation for a binary treatment: A tutorial. Statistics in Medicine. 2017

simmsm

R package that simulates data suitable for fitting Marginal Structural Model.

simulation-codes-for-causal-inference-for-polypharmacy

simulation-scenarios-jama

code for article in JAMA

simulationperformance

simulations-snns-vs-cox

This repository stores the R-code of a simulation study to compare survival neural networks (SNNs) with Cox models for clinical trial data. The predictive performance of ML techniques is compared with statistical models in a simple clinical setting (small/moderate sample size, small number of predictors) with Monte Carlo simulations. Synthetic data (250 or 1000 patients) are generated that closely resemble 5 prognostic factors pre-selected based on a European Osteosarcoma Intergroup study (MRC BO06/EORTC 80931). Comparison is performed between two partial logistic artificial neural networks (PLANN original by Biganzoli et al. 1998, Statistics in medicine, 17(10), 1169-1186 and PLANN extended by Kantidakis et al. 2020 BMC medical research methodology, 20(1), 1-14) as well as Cox models for 20, 40, 61, and 80% censoring. Survival times are generated from a log-normal distribution. Models are contrasted in terms of C-index, Brier score at 0-5 years, Integrated Brier Score (IBS) at 5 years, and miscalibration at 2 and 5 years. Endpoint of interest is overall survival. Note: PLANN original/extended are tuned based on IBS at 5 years and C-index.

spph504-007

SPPH 504 (section 007): Application of Epidemiological Methods

spph504007surveydata

Survey Data: Design and Examples

stability-confounder-select

Confounder selection strategies targeting stable treatment effect estimators

stanford_dl_ex

Programming exercises for the Stanford Unsupervised Feature Learning and Deep Learning Tutorial

stat-learning

Notes and exercise attempts for "An Introduction to Statistical Learning"

stratos-tg4-guidance-paper

Contains code for the worked examples in the TG4 guidance paper

stremr

Streamlined Estimation for Static, Dynamic and Stochastic Treatment Regimes in Longitudinal Data