train_and_predict.md |
train_and_predict.Rmd |
Generates LOO (leave-one-out) predictions for each state for which we have observed responses to one or both of the focal questions. |
train_and_predict_male_q66.md |
train_and_predict_male_q66.Rmd |
Same as train_and_predict but just for male respondents and the same-sex contact question. |
evaluate_loo_predictions.md |
evaluate_loo_predictions.Rmd |
Evaluates the LOO predictions against the true proportions we observe in YRBS data. |
make_final_state_predictions.md |
make_final_state_predictions.Rmd |
Combines the LOO predictions with true proportions and makes predictions for all the states without responses to the focal questions. |