I have a 404*34 dataset comprising of a dependent variable "Y" (temporally and spatially variable) and 33 predictor variables, for which the majority have values t