Table 1: Time-variant inefficiency model (Battese and coelli [12]).

Number of observations = 746 Number of countries = 75Observations per country: minimum = 3
Average = 9.9, maximum = 11

Log likelihood = 272.07096 Wald (8) = 14,540.41
Prob > = 0.0000

CoefficientsStandard error Confidence interval 95%
LowerUpper

−0.11980.0455−2.63000.0080−0.2089−0.0307
0.24570.13301.85000.0650−0.01490.5064
0.37670.18832.00000.04500.00770.7458
−0.00750.0015−5.13000.0000−0.0103−0.0046
0.02750.01112.49000.01300.00580.0492
0.05720.02162.65000.00800.01500.0995
−0.06050.0272−2.22000.0260−0.1138−0.0072
0.01060.00234.61000.00000.00610.0152
−0.00630.0030−2.11000.0350−0.0121−0.0004
8.81151.83664.80000.00005.211912.4111

0.20740.06263.31000.00100.08460.3302
0.06520.01165.59000.00000.04230.0880
−2.79460.2291−12.20000.0000−3.2437−2.3456
0.67350.35141.92000.0550−0.01531.3623

0.06110.01400.03900.0958
0.66230.07860.49620.7961
0.04050.01400.01310.0679
0.02060.00110.01840.0229