Dependent Variable: DIVPJ

Method: ML - Binary Logit (Quadratic hill climbing)

Date: 12/09/05 Time: 18:01

Sample: 1 3

Included observations: 2241

Convergence achieved after 4 iterations

Covariance matrix computed using second derivatives

Variable

Coefficient

Std. Error

z-Statistic

Prob.

ROA

30.09517

2.273120

13.23959

0.0000

DIVLAST

1.857351

0.104558

17.76380

0.0000

FCFA

0.067542

0.098179

0.687944

0.4915

C

-2.125779

0.107349

-19.80255

0.0000

Mean dependent var

0.487282

S.D. dependent var

0.499950

S.E. of regression

0.415783

Akaike info criterion

1.055234

Sum squared resid

386.7230

Schwarz criterion

1.065434

Log likelihood

-1178.390

Hannan-Quinn criter.

1.058958

Restr. log likelihood

-1552.618

Avg. log likelihood

-0.525832

LR statistic (3 df)

748.4565

McFadden R-squared

0.241031

Probability(LR stat)

0.000000

Obs with Dep=0

1149

Total obs

2241

Obs with Dep=1

1092