LSSVM Prediction using LS-SVMLab toolbox v1.8

I’m trying to forecast a time series of air passengers using LSSVM with the help of the LS-SVMLab toolbox v1.8 from http://www.esat.kuleuven.be/sista/lssvmlab/, specifically the NARX model function.

I would like to know if it is the correct function to use, rather than the simlssvm. Also, I would like to know how I can set a range and intervals to the model parameters, gamma and sig2.

Lastly, I’ve read that trial-and-error method was used to decide on the parameters based on the lowest MAE/MSE. That is because it gives different model parameters for every run? So I would be looking at the set of parameters with the lowest cost value returned?

Thank you in advance.