# Table 6 Estimation Results of the Ordered Probit Model

Dependent variable: category of toll elasticity (from 1 to 4)
Robust t-statistics

Excel | CSV

Variable Coefficient Std. error t-statistic P-value
Speed on the alternative road 0.032 0.010 3.298 0.001
Percentage of heavy vehicles on the alternative road –0.053 0.017 –4.233 0.000
Motorway section length 0.024 0.010 4.268 0.000
Tourist dummy –1.227 0.358 –3.340 0.001
Limit_1 0.919 0.862 1.214 0.226
Limit_2 2.393 0.920 3.014 0.003
Limit_3 3.666 0.962 3.814 0.000
Observations 52
Likelihood ratio-statistic 25.60 (critical value at 5% = 9.49)

Notes: The limit points are the estimates of the threshold coefficients of the distribution function. That is, if F(X'β) is the distribution function of the unobserved continuous latent variable, the ordered probit model implies that:

If F(X'β) ≤ Limit_1, then the dependent variable falls into category 1 (low elasticity).

If Limit_1≤ F(X'β) ≤ Limit_2, then the dependent variable falls into category 2 (middle-low elasticity).

If Limit_2 ≤ F(X'β) ≤ Limit_3, then the dependent variable falls into category 3 (middle-high elasticity).

If F(X'β) > Limit_3, then the dependent variable falls into category 4 (high elasticity).

Updated: Saturday, May 20, 2017