What is the meaning of negative value in Linear Discriminant Analysis coefficient?
Hello Muhammad,
The coefficients generated for independent variables (IVs) in a discriminant analysis are simply weights that help create a single score (linear composite) that maximally separates group means (group membership is the dependent variable). The sign of a coefficient will depend in part on how the groups are coded by the statistical software you have used, as well as on the sift of IVs you have included and how they are quantified.
A negative coefficient would be interpreted as indicating that, when the other IVs are held constant, and increase in the IV of interest would mean that the discriminant function score for a case is predicted to decrease. Therefore, the case would be somewhat less like a higher scored group (on the discriminant function scores) and more like a lower scored group). A decrease in the IV of interest would suggest that the case is more like a higher scored group and less like a lower scored group member.
Good luck with your work.
David explains your queation in detail.
If two-class discrimination, yoy set object y=1for class 1 and -=1 for class 2.
The obtained regression coefficients are the same role of the discriminant coefficients. You can image the meaning of the discriminant coefficients.
See my Springer book that include many samples of discriminant analysis.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs