Regression with more than one response (dependent) variable.
Use this tag to refer to cases where regression is used to model more than one response variable. Use multiple-regression when your question centers on cases with one response.
Special techniques may be used to regress more than one response variable onto a set of predictors. Examples:
- pls regression (or special forms of canonical-correlation analysis)
- manova
- Bayesian multivariate linear regression
- multivariate adaptive regression splines
- multivariate probit modeling
More info and references:
- Explain the difference between multiple regression and multivariate regression, with minimal use of symbols/math
- Wikipedia on multivariate statistics
- Examples in Stata by th UCLA Statistical Consulting Group
- Afifi, A., Clark, V., & May, S. (2004). Computer-Aided Multivariate Analysis (4th ed.). Boca Raton: Chapman & Hall/CRC.
- Alexopoulos, E. C. (2010). Introduction to Multivariate Regression Analysis. Hippokratia, 14(Suppl 1), 23–28. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3049417/.
- Maitra, R. (2013, March 7). Multivariate linear regression models. Iowa State University: Multivariate Statistical Methods. Retrieved from http://www.public.iastate.edu/~maitra/stat501/lectures/MultivariateRegression.pdf.
- Marden, J. I. (2013). Multivariate Statistics: Old school. University of Illinois at Urbana-Champaign. Retrieved from http://istics.net/pdfs/multivariate.pdf.
- Kemp-Benedict's implementation of multivariate linear regression in R
