Poisson gravity model:
Y ~ Poisson(Z[b(1)Y(-1)/Z(-1)+b(2)X(1)+b(3)X(2)]), 
where:
Y=OUT_MIG_1984
Z=1983-all empl.
Y(-1)=OUT_MIG_1983
Z(-1)=1982-all empl.
X(1)=Fraction > %change(1983-all empl.)
X(2)=1

Initial parameter values based on OLS:
b(1)=0.8694967
b(2)=0.0065235
b(3)=0.0021677
Newton iteration succesfully completed after 5 iterations
Last absolute parameter change = 1.11022E-016
Last percentage change of the likelihood = 0.00000E+000

Maximum likelihood estimation results:
b(.) ML estimate of b(.) (t-value) [p-value]
b(1)           0.8682957 (325.684) [0.00000]
b(2)           0.0081801  (35.324) [0.00000]
b(3)           0.0002114   (0.747) [0.45506]
[The two-sided p-values are based on the normal approximation]
t-value of Ho: b(1)=1:            -49.400
Log likelihood (*):    1.68168926739E+007
Sample size (n):                       75
(*) Only the part that depends on the parameters

Hausman test:
Test statistic: 3.767
Null distribution: Chi-square(3)
p-value = 0.28769
Significance levels:        10%         5%
Critical values:           6.25       7.81
Conclusions:             accept     accept

If the model is correctly specified then the maximum likelihood
parameter estimators b(1),..,b(3), minus their true values, times the
square root of the sample size n, are (asymptotically) jointly normally
distributed with zero mean vector and variance matrix:

 5.33094021E-04  1.03940711E-06 -4.95050101E-05 
 1.03940711E-06  4.02186419E-06 -2.19540795E-06 
-4.95050101E-05 -2.19540795E-06  6.00667720E-06 


Bootstrap parameter values:
b(1) = 0.8682957
b(2) = 0.0081801
b(3) = 0.0002114
Number of simulations = 1000


Simulated p-values:
P[b(1) < 0.868296] =  0.48800 
P[b(2) < 0.008180] =  0.49700 
P[b(3) < 0.000211] =  0.52400 
Joint p-value:        0.00400 

Simulated quantiles of b(1):
 5%: 0.8638083
10%: 0.8648002
15%: 0.8655051
20%: 0.8660430
25%: 0.8664829
30%: 0.8669144
35%: 0.8672761
40%: 0.8677096
45%: 0.8680308
50%: 0.8683775
55%: 0.8686475
60%: 0.8690086
65%: 0.8693053
70%: 0.8697393
75%: 0.8702039
80%: 0.8706444
85%: 0.8710884
90%: 0.8716480
95%: 0.8728442

Simulated quantiles of b(2):
 5%: 0.0077811
10%: 0.0078836
15%: 0.0079386
20%: 0.0079833
25%: 0.0080253
30%: 0.0080576
35%: 0.0080919
40%: 0.0081253
45%: 0.0081565
50%: 0.0081818
55%: 0.0082180
60%: 0.0082450
65%: 0.0082821
70%: 0.0083122
75%: 0.0083482
80%: 0.0083800
85%: 0.0084220
90%: 0.0084713
95%: 0.0085660

Simulated quantiles of b(3):
 5%: -0.0002733
10%: -0.0001645
15%: -0.0000884
20%: -0.0000389
25%: 0.0000098
30%: 0.0000456
35%: 0.0000875
40%: 0.0001170
45%: 0.0001519
50%: 0.0001848
55%: 0.0002290
60%: 0.0002732
65%: 0.0003159
70%: 0.0003605
75%: 0.0004008
80%: 0.0004499
85%: 0.0005104
90%: 0.0005789
95%: 0.0007007

Simulated p-value of the Hausman test: 0.24800