如题，鄙人的代码是
gmm回归：
/* Newey-West robust estimation */
proc model data=mkteq;
Parms b0 b1 b2 b3 b8 b9 b10 b11 b12 b13 b14 b15 b16 b17 b18 b19
b20 b21 b22 b23 b24 b25 b26;
synch=b0+b1*loganal+b2*logsize+b3*logvol+b8*year05+b9*year06+b10*year07+
b11*year08+b12*year09+b13*year10+b15*indb+b16*indc+b17*indd
+b18*inde+b19*indf+b20*indg+b21*indh+b22*indi+b23*indj+b24*indk+b25*indl
+b26*indm;
fit synch/gmm kernel=(bart, 3, 0);
instruments synch logsize logvol year05 year06 year07 year08 year09 year10
indb indc indd inde indf indg indh indi indj indk indl indm;
run;
quit;
2SLS：
proc model data=mkteq;
Parms b0 b1 b2 b3 b8 b9 b10 b11 b12 b13 b14 b15 b16 b17 b18 b19
b20 b21 b22 b23 b24 b25 b26;
synch=b0+b1*loganal+b2*logsize+b3*logvol+b8*year05+b9*year06+b10*year07+
b11*year08+b12*year09+b13*year10+b14*year11+b15*indb+b16*indc+b17*indd
+b18*inde+b19*indf+b20*indg+b21*indh+b22*indi+b23*indj+b24*indk+b25*indl
+b26*indm;
ENDOGENOUS loganal;
EXOGENOUS logsize logvol year05 year06 year07 year08 year09 year10
year11 indb indc indd inde indf indg indh indi indj indk indl indm;
fit synch/n2sls  vardef=N;
instruments logsize logvol year05 year06 year07 year08 year09 year10
year11 indb indc indd inde indf indg indh indi indj indk indl indm
volatility a_ratio;
run;
quit;
但是系数b1差距很大的说……
gmm得到的是 -32.2782，而2SLS得到的是 -0.20529

到底是怎么回事呢？
O(∩_∩)O谢谢