Friday , August 23 2019 Home / Lars P. Syll / Understanding regression assumptions

# Understanding regression assumptions by
Lars Pålsson Syll
My articles My siteMy booksAbout me
Follow on:Twitter
Summary:
Although most social scientists can recite the formal definitions of the various regression assumptions, many have little appreciation of the substantive meanings of these assumptions. And unless the meanings of these assumptions are understood, regression analysis almost inevitably will be a rigid exercise in which a handful of independent variables are cavalierly inserted into a standard linear additive regression and coefficients are estimated. Although such an exercise may occasionally produce results that are worth believing, it will do so only when an analyst is very lucky … This monograph was written to encourage students to … think of regression assumptions as a vital set of conditions the applicability of which must be explicitly analyzed each time regression

Topics:
Lars Pålsson Syll considers the following as important:

This could be interesting, too:

Lars Pålsson Syll writes What — if anything — do p-values test?

Lars Pålsson Syll writes Econometrics and the problem of unjustified assumptions

Lars Pålsson Syll writes Sloppy regression interpretations

Lars Pålsson Syll writes On the applicability of statistics in social sciences Although most social scientists can recite the formal definitions of the various regression assumptions, many have little appreciation of the substantive meanings of these assumptions. And unless the meanings of these assumptions are understood, regression analysis almost inevitably will be a rigid exercise in which a handful of independent variables are cavalierly inserted into a standard linear additive regression and coefficients are estimated. Although such an exercise may occasionally produce results that are worth believing, it will do so only when an analyst is very lucky … This monograph was written to encourage students to … think of regression assumptions as a vital set of conditions the applicability of which must be explicitly analyzed each time regression analysis is utilized.

I thought this book was great when I first read it 25 years ago.
I still do. Professor at Malmö University. Primary research interest - the philosophy, history and methodology of economics.