Chapter 2: Linear Regression Models
The most common of all regression models is the linear regression model, introduced in this chapter. This chapter also introduces the notation and language used in this book so a common foundation is laid for all readers for the upcoming study of generalized linear models: linear regression models are a special case of generalized linear models. We first define linear regression models and introduce the relevant notation and assumptions. We then describe least-squares estimation for simple linear regression models and multiple regression models. The use of the R functions to fit linear regression models is explained, followed by a discussion of the interpretation of linear regression models. Inference procedures are developed for the regression coefficients, followed by analysis of variance methods. We then discuss methods for comparing nested models, and for comparing non-nested models. Tools to assist in model selection are then described.
- Davison, A.C., Hinkley, D.V.: Bootstrap Methods and their Application. Cambridge University Press (1997)Google Scholar
- Harrell Jr, F.: Regression Modeling Strategies: With Applications to Linear Models, Logistic Models, and Survival Analysis. Springer (2001)Google Scholar
- Kunutsor, S.K., Powles, J.W.: The effect of ambient temperature on blood pressure in a rural West African adult population: A cross-sectional study. Cardiovascular Journal of Africa 21(1), 17–20 (2010)Google Scholar
- Singer, J.D., Willett, J.B.: Improving the teaching of applied statistics: Putting the data back into data analysis. The American Statistician 44(3), 223–230 (1990)Google Scholar
- Tibshirani, R.: Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society. Series B (Methodological) 58(1), 267–288 (1996)Google Scholar
- Woodward, M.: Epidemiology: study design and data analysis, second edn. Chapman & Hall/CRC, Boca Raton, FL (2004)Google Scholar