Regression and Other Stories - Tidyverse Examples

• Examples by chapters
  • 1 Introduction
  • 2 Data and measurement
  • 3 Some basic methods in mathematics and probability
  • 4 Generative models and statistical inference
  • 5 Simulation
  • 6 Background on regression modeling
  • 7 Linear regression with a single predictor
  • 8 Fitting regression models
  • 9 Prediction and Bayesian inference
  • 10 Linear regression with multiple predictors
  • 11 Assumptions, diagnostics, and model evaluation
  • 12 Transformations
  • 13 Logistic regression
  • 14 Working with logistic regression
  • 15 Other generalized linear models
  • 16 Design and sample size decisions
  • 17 Poststratification and missing-data imputation
  • 18 Causal inference basics and randomized experiments
  • 19 Causal inference using regression on the treatment variable
  • 20 Observational studies with all confounders assumed to be measured
  • 21 More advanced topics in causal inference
  • 22 Advanced regression and multilevel models
  • A Computing in R
• Examples alphabetically

This repository contains Tidyverse implementations of examples from Regression and Other Stories by Andrew Gelman, Jennifer Hill, and Aki Vehtari (2020).

• Book website
• Original code provided with book
  • Code website
  • GitHub repository

Tidyverse version by Bill Behrman.

Examples by chapters

1 Introduction

• ElectionsEconomy/
  • hibbs_tv.md - Predicting presidential vote share from the economy
• ElectricCompany/
  • electric_tv.md - Analysis of “Electric Company” data
• Peacekeeping/
  • peace_tv.md - Outcomes after civil war in countries with and without United Nations peacekeeping
• SimpleCausal/
  • causal_tv.md - Simple graphs illustrating regression for causal inference
• Helicopters/
  • helicopters_tv.md - Example data file for helicopter flying time exercise

2 Data and measurement

• HDI/
  • hdi_tv.md - Human Development Index - Looking at data in different ways
• Pew/
  • pew_tv.md - Miscellaneous analyses using raw Pew data
• HealthExpenditure/
  • healthexpenditure_tv.md - Discovery through graphs of data and models
• Names/
  • lastletters_tv.md - Last letters - Distributions of last letters of names of American babies
• Congress/
  • congress_plots_tv.md - Predictive uncertainty for congressional elections
• AgePeriodCohort/
  • births_tv.md - Age adjustment

3 Some basic methods in mathematics and probability

• Mile/
  • mile_tv.md - Trend of record times in the mile run
• Metabolic/
  • metabolic_tv.md - How to interpret a power law or log-log regression
• CentralLimitTheorem/
  • heightweight_tv.md - Illustrate central limit theorem and normal distribution
• Stents/
  • stents_tv.md - Stents - comparing distributions

4 Generative models and statistical inference

• Coverage/
  • coverage_tv.md - Example of coverage
• Death/
  • polls_tv.md - Proportion of American adults supporting the death penalty
• Coop/
  • riverbay_tv.md - Example of hypothesis testing
• Girls/

5 Simulation

• ProbabilitySimulation/
  • probsim_tv.md - Simulation of probability models
• Earnings/
  • earnings_bootstrap_tv.md - Bootstrapping to simulate the sampling distribution

6 Background on regression modeling

• Simplest/
  • simplest_tv.md - Linear regression with a single predictor
• Earnings/
  • earnings_regression_tv.md - Predict respondents’ yearly earnings using survey data from 1990
• PearsonLee/
  • heights_tv.md - The heredity of height. Published in 1903 by Karl Pearson and Alice Lee.
• FakeMidtermFinal/
  • simulation_tv.md - Fake dataset of 1,000 students’ scores on a midterm and final exam

7 Linear regression with a single predictor

• ElectionsEconomy/
  • hibbs_tv.md - Predicting presidential vote share from the economy
  • hibbs_coverage_tv.md - Checking the coverage of intervals
• Simplest/
  • simplest_tv.md - Linear regression with a single predictor

8 Fitting regression models

• ElectionsEconomy/
  • hills_tv.md - Present uncertainty in parameter estimates
  • hibbs_tv.md - Predicting presidential vote share from the economy
• Influence/
  • influence_tv.md - Influence of individual points in a fitted regression

9 Prediction and Bayesian inference

• ElectionsEconomy/
  • hibbs_tv.md - Predicting presidential vote share from the economy
  • bayes_tv.md - Demonstration of Bayesian information aggregation
• Earnings/
  • height_and_weight_tv.md - Predict weight
• SexRatio/
  • sexratio_tv.md - Example where an informative prior makes a difference

10 Linear regression with multiple predictors

• KidIQ/
  • kidiq_tv.md - Linear regression with multiple predictors
• Earnings/
  • height_and_weight_tv.md - Predict weight
• Congress/
  • congress_tv.md - Predictive uncertainty for congressional elections
• NES/
  • nes_linear_tv.md - Fitting the same regression to many datasets
• Beauty/
  • beauty_tv.md - Student evaluations of instructors’ beauty and teaching quality

11 Assumptions, diagnostics, and model evaluation

• KidIQ/
  • kidiq_tv.md - Linear regression with multiple predictors
• Residuals/
  • residuals_tv.md - Plotting the data and fitted model
• Introclass/
  • residual_plots_tv.md - Plot residuals vs. predicted values, or residuals vs. observed values?
• Newcomb/
  • newcomb_tv.md - Posterior predictive checking of Normal model for Newcomb’s speed of light data
• Unemployment/
  • unemployment_tv.md - Time series fit and posterior predictive model checking for unemployment series
• Rsquared/
  • rsquared_tv.md - Bayesian R^2
• CrossValidation/
  • crossvalidation_tv.md - Demonstration of cross validation
• FakeKCV/
  • fake_kcv_tv.md - Demonstration of K-fold cross-validation using simulated data
• Pyth/

12 Transformations

• KidIQ/
  • kidiq_tv.md - Linear regression with multiple predictors
• Earnings/
  • earnings_regression_tv.md - Predict respondents’ yearly earnings using survey data from 1990
• Gay/
  • gay_simple_tv.md - Simple models (linear and discretized age) and political attitudes as a function of age
• Mesquite/
  • mesquite_tv.md - Predicting the yields of mesquite bushes
• Student/
  • student_tv.md - Models for regression coefficients
• Pollution/
  • pollution_tv.md - Pollution data

13 Logistic regression

• NES/
  • nes_logistic_tv.md - Logistic regression, identifiability, and separation
• LogisticPriors/
  • logistic_priors_tv.md - Effect of priors in logistic regression
• Arsenic/
  • arsenic_logistic_building_tv.md - Building a logistic regression model: wells in Bangladesh
• Rodents/

14 Working with logistic regression

• LogitGraphs/
  • logitgraphs_tv.md - Different ways of displaying logistic regression
• Arsenic/
  • arsenic_logistic_building_tv.md - Building a logistic regression model: wells in Bangladesh
  • arsenic_logistic_apc_tv.md - Average predictive comparisons for a logistic regression model: wells in Bangladesh
  • arsenic_logistic_residuals_tv.md - Residual plots for a logistic regression model: wells in Bangladesh
• NES/
  • nes_logistic_tv.md - Logistic regression, identifiability, and separation

15 Other generalized linear models

• PoissonExample/
  • poisson_regression_tv.md - Demonstrate Poisson regression with simulated data
• Roaches/
  • roaches_tv.md - Analyze the effect of integrated pest management on reducing cockroach levels in urban apartments
• Storable/
  • storable_tv.md - Ordered categorical data analysis with a study from experimental economics, on the topic of “storable votes”
• Robit/
  • robit_tv.md - Comparison of robit and logit models for binary data
• Earnings/
  • earnings_compound_tv.md - Compound discrete-continuous model
• RiskyBehavior/
  • risky_tv.md Risky behavior data
• NES/
• Lalonde/
• Congress/
• AcademyAwards/

16 Design and sample size decisions

• SampleSize/
  • simulation_tv.md - Sample size simulation
• FakeMidtermFinal/
  • simulation_based_design_tv.md - Fake dataset of a randomized experiment on student grades
• ElectricCompany/
  • electric_tv.md - Analysis of “Electric Company” data

17 Poststratification and missing-data imputation

• Poststrat/
  • poststrat_tv.md - Poststratification after estimation
  • poststrat2_tv.md - Poststratification after estimation
• Imputation/
  • imputation_tv.md - Regression-based imputation for the Social Indicators Survey

18 Causal inference basics and randomized experiments

• Sesame/
  • sesame_tv.md - Causal analysis of Sesame Street experiment

19 Causal inference using regression on the treatment variable

• ElectricCompany/
  • electric_tv.md - Analysis of “Electric Company” data
• Incentives/
  • incentives_tv.md - Simple analysis of incentives data
• Cows/

20 Observational studies with all confounders assumed to be measured

• ElectricCompany/
  • electric_tv.md - Analysis of “Electric Company” data
• Childcare/
  • childcare_tv.md - Infant Health and Development Program (IHDP) example
• Lalonde/

21 More advanced topics in causal inference

• Sesame/
  • sesame_tv.md - Causal analysis of Sesame Street experiment
• ChileSchools/
  • chile_schools_tv.md - ChileSchools example.
• Bypass/

22 Advanced regression and multilevel models

• Golf/
  • golf_tv.md - Gold putting accuracy: Fitting a nonlinear model using Stan
• Gay/
  • gay_tv.md - Nonlinear models (LOESS and spline) and political attitudes as a function of age
• ElectionsEconomy/
  • hibbs_tv.md - Predicting presidential vote share from the economy
• Scalability/
  • scalability_tv.md - Demonstrate computation speed with 100,000 observations

A Computing in R

• Coins/
• Mile/
  • mile_tv.md - Trend of record times in the mile run
• Earnings/
  • earnings_data_tv.md - Read in and prepare earnings data
• Parabola/
  • parabola_tv.md - Demonstration of using Stan for optimization
• Restaurant/
  • restaurant_tv.md - Demonstration of using Stan for optimization
• DifferentSoftware/
  • linear_tv.md - Linear regression using different software options

Examples alphabetically

• AcademyAwards/
• AgePeriodCohort/
  • births_tv.md - Age adjustment
• Arsenic/
  • arsenic_logistic_building_tv.md - Building a logistic regression model: wells in Bangladesh
  • arsenic_logistic_apc_tv.md - Average predictive comparisons for a logistic regression model: wells in Bangladesh
  • arsenic_logistic_residuals_tv.md - Residual plots for a logistic regression model: wells in Bangladesh
• Beauty/
  • beauty_tv.md - Student evaluations of instructors’ beauty and teaching quality
• Bypass/
• CentralLimitTheorem/
  • heightweight_tv.md - Illustrate central limit theorem and normal distribution
• Childcare/
  • childcare_tv.md - Infant Health and Development Program (IHDP) example
• ChileSchools/
  • chile_schools_tv.md - ChileSchools example.
• Coins/
• Congress/
  • congress_tv.md - Predictive uncertainty for congressional elections
  • congress_plots_tv.md - Predictive uncertainty for congressional elections
• Coop/
  • riverbay_tv.md - Example of hypothesis testing
• Coverage/
  • coverage_tv.md - Example of coverage
• Cows/
• CrossValidation/
  • crossvalidation_tv.md - Demonstration of cross validation
• Death/
  • polls_tv.md - Proportion of American adults supporting the death penalty
• DifferentSoftware/
  • linear_tv.md - Linear regression using different software options
• Earnings/
  • earnings_bootstrap_tv.md - Bootstrapping to simulate the sampling distribution
  • earnings_compound_tv.md - Compound discrete-continuous model
  • earnings_regression_tv.md - Predict respondents’ yearly earnings using survey data from 1990
  • height_and_weight_tv.md - Predict weight
  • earnings_data_tv.md - Read in and prepare earnings data
• ElectionsEconomy/
  • bayes_tv.md - Demonstration of Bayesian information aggregation
  • hibbs_coverage_tv.md - Checking the model-fitting procedure using fake-data simulation.
  • hibbs_tv.md - Predicting presidential vote share from the economy
  • hills_tv.md - Present uncertainty in parameter estimates
• ElectricCompany/
  • electric_tv.md - Analysis of “Electric Company” data
• FakeKCV/
  • fake_kcv_tv.md - Demonstration of K-fold cross-validation using simulated data
• FakeMidtermFinal/
  • simulation_tv.md - Fake dataset of 1,000 students’ scores on a midterm and final exam
  • simulation_based_design_tv.md - Fake dataset of a randomized experiment on student grades
• Gay/
  • gay_simple_tv.md - Simple models (linear and discretized age) and political attitudes as a function of age
  • gay_tv.md - Nonlinear models (LOESS and spline) and political attitudes as a function of age
• Girls/
• Golf/
  • golf_tv.md - Gold putting accuracy: Fitting a nonlinear model using Stan
• HDI/
  • hdi_tv.md - Human Development Index - Looking at data in different ways
• HealthExpenditure/
  • healthexpenditure_tv.md - Discovery through graphs of data and models
• Helicopters/
  • helicopters_tv.md - Example data file for helicopter flying time exercise
• Imputation/
  • imputation_tv.md - Regression-based imputation for the Social Indicators Survey
• Incentives/
  • incentives_tv.md - Simple analysis of incentives data
• Influence/
  • influence_tv.md - Influence of individual points in a fitted regression
• Introclass/
  • residual_plots_tv.md - Plot residuals vs. predicted values, or residuals vs. observed values?
• KidIQ/
  • kidiq_tv.md - Linear regression with multiple predictors
• Lalonde/
• LogisticPriors/
  • logistic_priors_tv.md - Effect of priors in logistic regression
• LogitGraphs/
  • logitgraphs_tv.md - Different ways of displaying logistic regression
• Mesquite/
  • mesquite_tv.md - Predicting the yields of mesquite bushes
• Metabolic/
  • metabolic_tv.md - How to interpret a power law or log-log regression
• Mile/
  • mile_tv.md - Trend of record times in the mile run
• Names/
  • lastletters_tv.md - Last letters - Distributions of last letters of names of American babies
• NES/
  • nes_linear_tv.md - Fitting the same regression to many datasets
  • nes_logistic_tv.md - Logistic regression, identifiability, and separation
• Newcomb/
  • newcomb_tv.md - Posterior predictive checking of Normal model for Newcomb’s speed of light data
• Parabola/
  • parabola_tv.md - Demonstration of using Stan for optimization
• Peacekeeping/
  • peace_tv.md - Outcomes after civil war in countries with and without United Nations peacekeeping
• PearsonLee/
  • heights_tv.md - The heredity of height. Published in 1903 by Karl Pearson and Alice Lee.
• Pew/
  • pew_tv.md - Miscellaneous analyses using raw Pew data
• PoissonExample/
  • poisson_regression_tv.md - Demonstrate Poisson regression with simulated data
• Pollution/
  • pollution_tv.md - Pollution data
• Poststrat/
  • poststrat_tv.md - Poststratification after estimation
  • poststrat2_tv.md - Poststratification after estimation
• ProbabilitySimulation/
  • probsim_tv.md - Simulation of probability models
• Pyth/
• Residuals/
  • residuals_tv.md - Plotting the data and fitted model
• Restaurant/
  • restaurant_tv.md - Demonstration of using Stan for optimization
• RiskyBehavior/
  • risky_tv.md Risky behavior data
• Roaches/
  • roaches_tv.md - Analyze the effect of integrated pest management on reducing cockroach levels in urban apartments
• Robit/
  • robit_tv.md - Comparison of robit and logit models for binary data
• Rodents/
• Rsquared/
  • rsquared_tv.md - Bayesian R^2
• SampleSize/
  • simulation_tv.md - Sample size simulation
• Scalability/
  • scalability_tv.md - Demonstrate computation speed with 100,000 observations
• Sesame/
  • sesame_tv.md - Causal analysis of Sesame Street experiment
• SexRatio/
  • sexratio_tv.md - Example where an informative prior makes a difference
• SimpleCausal/
  • causal_tv.md - Simple graphs illustrating regression for causal inference
• Simplest/
  • simplest_tv.md - Linear regression with a single predictor
• Stents/
  • stents_tv.md - Stents - comparing distributions
• Storable/
  • storable_tv.md - Ordered categorical data analysis with a study from experimental economics, on the topic of “storable votes”
• Student/
  • student_tv.md - Models for regression coefficients
• Unemployment/
  • unemployment_tv.md - Time series fit and posterior predictive model checking for unemployment series