$ n=31;
$ t=21;
$ k=3;
$load y[t,n]=theil.txt;
$ load x[t,n*k]=x.txt;
$ library pcoint coint pgraph;
$ _ker_fun=&fejer;
$ {fm_res,b,c,d,e}=fm_rd(y,x);
The Fully-Modified Estimators
The FM_beta1 is:  1.3548
The t-ratio1 is: 27.3145            probabiltiy(t)=  0.0000 probability(N)= 0.0000
The FM_beta2 is: -1.9581
The t-ratio2 is: -18.1829            probabiltiy(t)=  0.0000 probability(N)= 0.0000
The FM_beta3 is:  1.2223
The t-ratio3 is:  8.1109            probabiltiy(t)=  0.0000 probability(N)= 0.0000
R square is: -0.6819
Adjusted Rsquare is: -0.6897

R2为负，这是怎么回事
郁闷了
The Conventional T test and OLS estimatorThe beta1 is:  0.4748
The t-ratio1 is: 13.5293            probabiltiy(t)=  0.0000 probability(N)= 0.0000
The beta2 is: -0.2553
The t-ratio2 is: -2.7611            probabiltiy(t)=  0.0030 probability(N)= 0.0029
The beta3 is:  0.3675
The t-ratio3 is:  3.4979            probabiltiy(t)=  0.0003 probability(N)= 0.0002
R square is:  0.4946
Adjusted Rsquare is:  0.4923

THE ADJUSTED T RATIO and VALUES
The adjusted beta1 is:  0.5830
The adjusted t-ratio1 is: 12.3414            probabiltiy(t)=  0.0000 probability(N)= 0.0000
The adjusted beta2 is: -0.2828
The adjusted t-ratio2 is: -2.7572            probabiltiy(t)=  0.0030 probability(N)= 0.0029
The adjusted beta3 is:  0.1986
The adjusted t-ratio3 is:  1.3841            probabiltiy(t)=  0.0834 probability(N)= 0.0832
The bias of beta 1 is: -2.2721
The bias of beta 2 is:  0.5766
The bias of beta 3 is:  3.5451
R square is:  0.4946
Adjusted Rsquare is:  0.4843
$
The Dynamic OLS EstimatorsThe DOLS_beta1 is:  0.4016
The t-ratio1 is:  6.8826            probabiltiy(t)=  0.0000 probability(N)= 0.0000
The DOLS_beta2 is: -1.2375
The t-ratio2 is: -9.7677            probabiltiy(t)=  0.0000 probability(N)= 0.0000
The DOLS_beta3 is: -0.2130
The t-ratio3 is: -1.2015            probabiltiy(t)=  0.1151 probability(N)= 0.1148
R square is:  0.2758
Adjusted Rsquare is: -1.4901