I guess I just don't understand the wald concept ...at all

where u~N(0,sigma^2I), y and u are n*1,x1 is n*k1 and x2 is n*k2.b1 is k1*1 and b2 is k2*1.(x1 and x2 are non-stochastic)

Test H0:b2 = 0, 2 ways

A)estimate the unrestricted model, and formulate a Wald test statistic based on the linear restrictions RB=r.



and then use the residuals from the regresssion, namely u1*, as the regressor and regress U1* against x2. i.e. u1*=x2z2+v. Test the H0:z2=0

i) Formulate the Wald statistic from A
ii)formulate the wald statstic from b and make a comparison of i)

I really don't know what to do here....