Is a regression estimator consistent without an intercept

Statistics pp 459-496 | Cite as

  • Ludwig Fahrmeir
  • Rita artist
  • Iris Pigeot
  • Gerhard Tutz
Part of the Springer textbook book series (SLB)

Summary

Section 3.6 deals with how the influence of an explanatory feature X on a target feature Y can be represented and exploratively investigated. Both features are assumed to be scaled metrically, and it is assumed that the relationship between Y and X can be described by an approximate relation of the form \ (Y = f (X) + \ varepsilon \). It is f a deterministic regression function and ε a bug caused by X alone cannot be explained. The best known is the linear single regression, in which a linear regression function \ ([f = (X) = \ alpha + \ beta X \) is used as the “best-fit line”. We also consider the CAP model as an example in this chapter Y = “Stock return minus interest” and X = “Market return” and the rent index regression with Y = “Net rent” (or “net rent / sqm”) and X = “Living space”. Figures 3.17 and 3.20 from Section 3.6 show the scatter diagrams and the associated best-fit line.

This is a preview of subscription content, log in to check access.

Preview

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • Ludwig Fahrmeir
  • Rita artist
  • Iris Pigeot
  • Gerhard Tutz
  1. 1. Institute for Statistics, University of Munich, Munich, Germany
  2. 2. Institute for Statistics, University of Munich, Munich, Germany