STATISTICS FOR ENVIRONMENTAL STUDIES

博士前期課程地球環境学研究科

MGGE8060

コース情報

担当教員: 中川善典

単位数: 2

年度: 2024

学期: 春学期

曜限: 水1

形式: 対面授業

レベル: 600

アクティブラーニング: なし

他学部履修: 可

評価方法

レポート

10%

定期試験

定期試験期間中

90%

詳細情報

概要

In order to read and conduct research in environmental studies, whether it be in the natural sciences or the social sciences, it is important to have an understanding of statistics. Having deeply acknowledged this fact, one should then proceed to acquire various concepts fundamental to statistics. Subsequently, one should learn how to conduct data analysis using multiple regression analysis, panel data analysis, and instrumental variable methods. The utilization of statistical software R is adopted for this purpose.

目標

You will become able to conduct data analysis using multiple regression analysis, panel data analysis, and instrumental variable methods.

授業外の学習

To ensure the consolidation of the content learned in lectures, homework assignments will be given after each lecture as a general rule. Additionally, individuals who have not studied statistics will engage in preparatory learning prior to the commencement of the course using supplementary materials distributed separately.

所要時間: 190 minutes

スケジュール

Simple Regression Analysis 1: Least Squares Method
Simple Regression Analysis 2: What is the error term in a regression model?
Simple Regression Analysis 3: What does it mean to view the estimated values of regression coefficients as random variables?
Simple Regression Analysis 4: Dummy variables, standardized regression coefficients, coefficient of determination
Multiple Regression Analysis 1: Correlation = Spurious correlation + Indirect effect + Direct effect
Multiple Regression Analysis 2: Difference between P(Y|X) and P(Y|do(X))
Multiple Regression Analysis 3: Backdoor criteria as a condition for eliminating spurious correlation and representing the overall effect by partial regression coefficients
Multiple Regression Analysis 4: Execution of multiple regression analysis using statistical software R
Panel Data Analysis 1: First difference method as a solution when the backdoor criteria cannot be met
Panel Data Analysis 2: Mean difference method as a solution when the backdoor criteria cannot be met
Panel Data Analysis 3: Execution of panel data analysis using statistical software R
Instrumental Variable Method 1: Instrumental variable method as a solution when the backdoor criteria cannot be met (in the case of a simple regression model)
Instrumental Variable Method 2: Instrumental variable method as a solution when the backdoor criteria cannot be met (in the case of a multiple regression model)
Instrumental Variable Method 3: Execution of instrumental variable method using statistical software R

教科書

No textbooks will be specified.

参考書

書籍情報はありません。