Postcalibration sensitivity procedure for regressor variable errors
- 1 October 1986
- journal article
- research article
- Published by Canadian Science Publishing in Canadian Journal of Forest Research
- Vol. 16 (5) , 1120-1123
- https://doi.org/10.1139/x86-195
Abstract
Random errors in the independent variables of a calibrated model can make predictions less precise. A method is given for approximating the precision of predictions when there are random errors in the postcalibration data. An example reveals that when independent variables are not error free, an "optimal" model can yield worse predictions than a "suboptimal" model.This publication has 0 references indexed in Scilit: