VECM restrictions

[keanodahdemon]

Dear Sirs,
I have a problem regarding imposing restrictions on the VECM. I'm doing a test on the unbiasedness hypothesis, more specifically testing the restrictions B(1,1)=0 and B(1,2)= -1 in the cointegrating relationship B'Xt-1 = ( 1 B1 B2) (St-1 1 Ft-1;t-n-1)'. If these restrictions hold, then the price of a futures contract is an unbiased predictor of the realized spot price. When I impose this in Eviews 7.0 I got results that seemed biased (I expect the hypothesis to be true). I contacted a professor, and he tells me that Eviews can give unbiased result when imposing these restrictions simultaneously (after what I can see myself is that restriction nr.2 is not possible to perfrom). Can anyone give me a clear answer on this? And maybe a way to circumvent this problem.