Thanks Allin, that clinches it:

Read datafile /usr/local/share/gretl/data/misc/denmark.gdt

Model 1: OLS, using observations 1975:1-1987:3 (T = 51)
Dependent variable: Y

             coefficient    std. error    t-ratio   p-value 
  ----------------------------------------------------------
  const       0.573197      0.457911       1.252    0.2179  
  time        0.000288957   0.000360372    0.8018   0.4274  
  Y_1         0.800428      0.153400       5.218    5.90e-06 ***
  Y_2         0.120277      0.198738       0.6052   0.5485  
  Y_3        −0.0469882     0.180931      −0.2597   0.7964  
  Y_4        −0.213941      0.144448      −1.481    0.1464  
  X           0.336480      0.104987       3.205    0.0027   ***
  X_1        −0.0594923     0.157019      −0.3789   0.7068  
  X_2        −0.430368      0.150459      −2.860    0.0067   ***
  X_3         0.0952069     0.161941       0.5879   0.5599  
  X_4         0.181183      0.135096       1.341    0.1874  

Mean dependent var   5.959878   S.D. dependent var   0.069523
Sum squared resid    0.015477   S.E. of regression   0.019670
R-squared            0.935959   Adjusted R-squared   0.919949
F(10, 40)            58.46033   P-value(F)           1.11e-20
Log-likelihood       134.1900   Akaike criterion    −246.3801
Schwarz criterion   −225.1300   Hannan-Quinn        −238.2598
rho                 −0.024300   Durbin-Watson        2.002194

Excluding the constant, p-value was highest for variable 10 (Y_3)

C

On 9 December 2016 at 03:47, Allin Cottrell <cottrell@wfu.edu> wrote:
On Fri, 9 Dec 2016, Clive Nicholas wrote:

Thanks Jack,

The reformatting works, but then chokes on

varsim(Y, ARbeta, lX, Xbeta, lD, Dbeta, e,
The symbol 'e' is undefined

Error executing script: halting
varsim(Y, ARbeta, lX, Xbeta, lD, Dbeta, e, p, &ysim)

And indeed 'e' is not defined...

 You're right, Artur's example does not run as posted. It's not totally obvious to me where the missing "e" series is supposed to come from, though maybe one should insert

series e = resample($uhat)

after the ols command?

Allin Cottrell

_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users



--
Clive Nicholas

"My colleagues in the social sciences talk a great deal about methodology. I prefer to call it style." -- Freeman J. Dyson