Code: Select all
x = @nan(x,y)
However, if we wanted to apply more rigorous pre-processing, then we may consider something like knn imputation. Or even better, something like this: https://www.researchgate.net/publication/271710133_Balanced_k-nearest_neighbor_imputation. Balanced knn imputation has some great stochastic properties, but it seems a little difficult to implement via eviews programming (at least for me).