Machine Learning for Economics and Business
Preface
Who Should Read This Book
How to Use This Book
A Note on References
1
Randomized Controlled Trial, A/B/N Testing, and Multi-Armed Bandit Algorithms
1.1
Introduction
1.2
The Explore-Exploit Tradeoff
1.3
Epsilon Greedy
1.4
Optimistic Initial Values
1.5
Upper Confidence Bound (UCB)
1.6
Gradient Bandit Algorithm
1.7
Thompson Sampling (Bayesian Bandits)
1.8
Conjugate Prior
1.9
Thompson Sampling: Code
1.10
Comparing the Algorithms
1.11
Summary and Extensions
1.12
References
2
Discrete Choice, Classification, and Tree-Based Ensemble Algorithms
2.1
Introduction
2.2
The Bias-Variance Tradeoff
2.3
Decision Tree
2.4
Split Criterion
2.5
Pruning
2.6
Bagging and Random Forest
2.7
Boosting and AdaBoost
2.8
Gradient Boosting and XGBoost
2.9
Python Implementation with scikit-learn
2.10
Confusion Matrix and other Performance Metrics
2.11
Comparison the Algorithms
2.12
Summary
2.13
References
3
Time Series, Forecasting, and Deep Learning Algorithms
3.1
Introduction
3.2
Time Series Implementation in
statsmodels
3.3
Artificial Neural Network (ANN)
3.4
Recurrent Neural Network (RNN)
3.5
Convolutional Neural Network (CNN)
3.6
Deep Learning Algorithms in TensorFlow/Keras
3.7
Facebookâs Prophet
3.8
Summary
3.9
References
4
Regression Reconsidered
5
Causal Inference Reconsidered
6
More than Meets the Eye
Published with bookdown
Machine Learning for Economics and Business
Chapter 5
Causal Inference Reconsidered