T. Furukawa, T. Hirai, T. Kitamura, K. Takahashi: Application of the Center for Epidemiologic Studies Depression
Scale among first-visit psychiatric patients: a new approach to
improve its performance. Journal of Affective Disorders 46 (1997) 1–13
Even though we warned that causality is never proven by regression analysis, when a researcher specifies that Y=f(X), an implicit causal linkage is assumed, In general, this functional relationship runs from X to Y (i.e., the value of Y is dependent on the value of X). But in many situations the dependency may run both ways (i.e., X is also a function of Y). [Schroeder, Sjoquest, Stephan (1986) Understanding Regression Analysis: An Introductory Guide (Quantitative Applications in the Social Sciences), p77, Sage Publications]
(略)…just because a regression analysis has indicated a strong relationship between two variables, this does not imply that the variables are related in any causal sense. Causality implies necessary correlation. Regression analysis can only address the issues on correlation. It cannot address the issue of necessity. Thus, our expectations of discovering cause and effect relationships from regression should be modest.[Montgomery, Peck, & Vining (2001) Introduction to Linear Regression Analysis, 3rd Edition, pp42, Wiley-Interscience]
以下の文献は、回帰分析が因果関係を証明できないということを証明している研究です。
Glymour, Spirtes & Schenines(1994) In Place of Regression in Humphreys (Ed) Patrick Suppes: Scientific Philosopher: Volume 1: Probability and Probabilistic Causality , 339-366, Springer
とりあえず、ネット上の情報も。↓
http://www.csus.edu/indiv/y/yangy/145Ch3.htm
4.2 Regression Analysis
(省略)The primary objective in regression analysis is to estimate and/or predict the (population) mean or expected value of the dependent variable in terms of the known or fixed values of the independent variable(s). But dependence of one variable on other variable(s) in regression analysis does not necessarily mean causation. As Kendall and Stuart point out, "A statistical relationship, however strong and however suggestive, can never establish causal connection: our ideas of causation must come from outside statistics, ultimately from some theory or other."