Modèle Gaussien

Avec Glim4:

glim
$pag on   !pour avoir la pagination
$echo on  !pour avoir le detail des fichiers importes
$uni 5$
$dat x1 x2 y$
$read
1  2  33.11
2  1  2.5
3  0  0.4
4  1  0.1
5  2  0.2
$yva y$
$fit x1+x2+x1*x2$
   deviance =  59.699
residual df =   1
$loo %x2$
59.70
$fit -x1.x2$
   deviance =  138.72 (change =   +79.02)
residual df =    2    (change =    +1   )
$loo %x2$
138.7

$dis a$
          estimate        s.e.     parameter
     1       16.74       10.58      1
     2      -6.822       2.634      X1
     3       9.160       4.977      X2
scale parameter 69.36

$dis r$
   unit   observed    fitted   residual
      1    33.1100   28.2340      4.876
      2     2.5000   12.2520     -9.752
      3     0.4000   -3.7300      4.130
      4     0.1000   -1.3920      1.492
      5     0.2000    0.9460     -0.746

$dis c$
correlations between parameter estimates
  1   1.0000
  2  -0.7467   1.0000
  3  -0.5644   0.0000   1.0000
         1        2        3

$dis v$      
(co)variance matrix of parameter estimates
  1       112.0
  2      -20.81       6.936
  3      -29.72       0.000       24.77
            1           2           3

$extract %di$   

$loo  %di$   
       %DI
 1   23.7754
 2   95.1015
 3   17.0569
 4    2.2261
 5    0.5565

$stop