cnchapman / choicetools

Tools for Choice Modeling, Conjoint Analysis, and MaxDiff analysis of Best-Worst Surveys
Apache License 2.0
27 stars 14 forks source link

SSE #3

Open MikeRebel opened 3 years ago

MikeRebel commented 3 years ago

Is this SSE calculation in generateMNLrandomTab() correct? Calculations now:

attrLevels <- c(Brand=4, Style=4, Price=5, Color=2, Size=4)
att.base <- 1/rep(attrLevels, attrLevels)
new.mss <- sum((colMeans(new.des) - att.base)^2)

SSE is изображение but att.base is calculated in strange way. 1 devide by attribute level count in attrLevels.

att.base Brand Brand Brand Brand Style Style Style Style Price Price Price Price Price Price 0.2500000 0.2500000 0.2500000 0.2500000 0.2500000 0.2500000 0.2500000 0.2500000 0.1666667 0.1666667 0.1666667 0.1666667 0.1666667 0.1666667 Color Color Size Size Size Size 0.5000000 0.5000000 0.2500000 0.2500000 0.2500000 0.2500000 should be изображение