Gradients and Lagrange Multiplier Tests

[psantoro]

Hello,

I'm trying to test for linearity against STAR models via lagrange multiplier procedure of Luukkonen, et. al. (1988). When estimating a VAR eviews has menu options for Wald coefficien restriction tests and Logl-type procedures, but none for an LM test. In reading the manual I found that calculating and grouping the gradients can be done when doing the type of thing I'm trying to do. I don't quite yet understand the relationship or how I would proceed. Could someone elaborate for me?