RT Article
T1 Working with Misspecified Regression Models
JF Journal of quantitative criminology
VO 34
IS 3
SP 633
OP 655
A1 Berk, Richard
A2 Brown, Lawrence 1940-
A2 Buja, Andreas
A2 George, Edward
A2 Zhao, Linda
LA English
YR 2018
UL https://krimdok.uni-tuebingen.de/Record/1689132604
AB Conventional statistical modeling in criminology assumes proper model specification. Very strong and unrebutted criticisms have existed for decades. Some respond that although the criticisms are correct, there is for observational data no alternative. In this paper, we provide an alternative.
K1 Reproducibility
K1 Misspecification
K1 Regression
DO 10.1007/s10940-017-9348-7