These notebooks are used to generate the figures of the paper: STEP-BY-STEP CAUSAL ANALYSIS OF EHRS TO GROUND DECISION MAKING :
0_description_albumin_for_sepsis.py: Description of the covariates for the population (Tables 2 and apd:7).1_sensitivity_albumin_for_sepsis_report.py: Sensitivity analysis of the causal effect of albumin for sepsis. We vary the causal and statistical estimators as well as the aggregations of the features (Figures 7 and apd:13: Forest plot for the vibration analysis).2_immortal_bias_albumin_for_sepsis_report: Sensitivity analysis for the immortal time bias. We vary the blank period between inclusion and treamtent allocation(Figures 8 and apd:14).3_feature_agg_sensitivity_albumin_for_sepsis_report.py: Sensitivity analysis for the aggregation of the features (Figure apd:16).4_cate_albumin_for_sepsis_report.py: Sensitivity analysis for the heterogeneity of the treatment (Figures 9, apd:17 and apd:18).5_prediction_failure_report.py: Motivating example for the prediction failure (Figure apd:10).