Talk:Linear predictor function

The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

In the Basic Form section, the matrix vector definitions shouldn't have dimensions (p+1)-by-1 and 1-by-(p+1) ? — Preceding unsigned comment added by ZMNunes (talk • contribs) 15:51, 9 July 2013‎

Proposed merge with Linear model

A linear model is a model with a linear predictor function. These articles discuss the same concept, but the article linear model is mostly restricted to regression. QVVERTYVS (hm?) 11:13, 5 January 2014 (UTC)[reply]

Because there has been no discussion in three years, the proposed merge will not be implemented. Will remove the merge-proposal banners from the two articles. NewYorkActuary (talk) 06:05, 18 January 2017 (UTC)[reply]

Mistake? Length of p or n?

The basic form of a linear predictor function f(i) for data point i (consisting of p explanatory variables), for i = 1, ..., n, is

f(i) = \beta_0 + \beta_1 x_{i1} + \cdots + \beta_p x_{ip},

should the n be rewritten as p? should the final x_{ip} be x_{in}?

Looking at it I think it should be

f(i) = \beta_0 + \beta_1 x_{i1} + \cdots + \beta_p x_{in}

let p = n + 1

then β is of size p (not p+1), x grows to size p after prepending x_{i0} = 1, and everything else works out — Preceding unsigned comment added by 199.223.21.100 (talk) 17:25, 21 March 2016 (UTC)[reply]