大学院
HOME 大学院 Machine Learning for Economics
戻る
学内のオンライン授業の情報漏洩防止のため，URLやアカウント、教室の記載は削除しております。
最終更新日：2024年4月22日

授業計画や教室は変更となる可能性があるため、必ずUTASで最新の情報を確認して下さい。
UTASにアクセスできない方は、担当教員または部局教務へお問い合わせ下さい。

Machine Learning for Economics

The objective of this course is to study machine learning methods and their practical applications in empirical economic research. The machine learning methods covered in this course include regression and classification trees, bagging, random forests, boosting, neural network learning (deep learning), and bandit algorithms. We also study how these methodologies can be applied to statistical causal inference, public policy making, and macroeconomic forecasting. We additionally discuss utilization of alternative data, such as image data and text data, in economic research.
MIMA Search
時間割／共通科目コード
コース名
教員
学期
時限
291324-10
GEC-EC6322L3
Machine Learning for Economics
坂口　翔政
A1 A2
水曜4限
マイリストに追加
マイリストから削除
講義使用言語
英語
単位
2
実務経験のある教員による授業科目
NO
他学部履修
可
開講所属
経済学研究科
授業計画
1. Introduction and regression 2. Regression trees and classification trees 3. Ensemble Learning: Bagging, random forest, and boosting 4. Neural network learning (1): Neural network model, learning methods (stochastic gradient descent, backpropagation etc.) 5. Neural network learning (2): Convolutional neural network, image data analysis 6. Neural network learning (3): Recursive neural network, long-short time memory model, sequential data analysis (time series data, text data), sentiment analysis, macroeconomic forecasting 7. Basics of statistical causal inference 8. Applications of machine learning methods to causal inference (1) :Causal Forest 9. Applications of machine learning methods to causal inference (2): Double/debiased machine learning 10. Applications of machine learning methods to public policy making 11. Bandit problems and algorithms 12-13. Students' presentations The lecture plan might be changed.
授業の方法
Lectures and demo programs
成績評価方法
The course evaluation is based on two problem sets (30%), one student presentation (35%), and one term paper (35%).
教科書
NA
参考書
To be announced at the first lecture
履修上の注意
- Students are assumed to have the basic knowledge of probability and statistics. - Programming experience is useful but not required. - Having knowledge of statistical causal inference is useful.
戻る