VECM - dropping insignificant 1st difference regressors

[eviewsbeg]

Hello,

I was wondering if there is a way to drop the insignificant lagged first-difference regressors in the short-term equations of the VECM (but keep the corresponding variables in the cointegrating equations)? It seems that restrictions on the c(i,j) coefficients are not accepted, so is there an alternative?

Thanks in advance.
EviewsBeg