CARIFORUM and UK EPA Study

æ ç

ö

p

ln

;

Z

=

therefore, ÷ è - ø Dependent Variable: TRDL Method: ML - Binary Logit (Newton-Raphson / Marquardt steps) Date: 09/23/21 Time: 17:44 Sample: 1 182 Included observations: 182 Convergence achieved after 8 iterations Coefficient covariance computed using observed Hessian 1 p

or the odds of occurrences, which can be converted into probabilities

Variable

Coefficient

Std. Error

z-Statistic Prob.

LFG

0.307140 0.290216 1.058315 0.2899 0.016663 0.056950 0.292592 0.7698 1.970226 1.049311 1.877639 0.0604 -3.093976 5.460230 -0.566638 0.5710 -4.834807 8.151617 -0.593110 0.5531 -0.101233 0.070042 -1.445314 0.1484 0.123234 0.163626 0.753144 0.4514 0.385957 0.074639 5.170982 0.0000 -13.70223 9.257715 -1.480088 0.1388

K

INFL

DI

EXR

GDPG

INT

EXP01

C

McFadden R-squared 0.670680 Mean dependent var

0.296703 0.221743 8.506375 -36.44389 72.88777 -110.6642 -0.200241

S.D. dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter.

0.458065 S.E. of regression 0.499383 Sum squared resid 0.657823 Log likelihood 221.3284 Restr. log likelihood 148.4406 Avg. log likelihood 0.563613 Deviance

Restr. deviance

LR statistic

Prob(LR statistic)

0.000000

Obs with Dep=0 Obs with Dep=1

128 Total obs

182

54

Only exports show a strong probability of increasing the average value of trade. Interpretation of the coefficient values is complicated by the fact that estimated coefficients from a binary model cannot be interpreted as the marginal effect on the dependent variable. The LR statistic tests the joint null hypothesis that all slope coefficients except the constant are zero and is computed as. This statistic, which is only reported when you include a constant in your specification, is used to test the overall significance of the model. The degrees of freedom is one less than the number of coefficients in the equation, which is the number of restrictions under test.

Model Results Incorporating the PCI

Page 224 of 241