Research Article

Investigating Travel Flow Differences between Peak Hours with Spatial Model with Endogenous Weight Matrix Using Automatic Vehicle Identification Data

Table 4

Estimation results of the SARBP-EWM model and binary probit model for weekday and weekend.

Explanatory variablesSARBP-EWM (weekday)SARBP-EWM (weekend)Binary probit model (weekday)Binary probit model (weekend)
Coef.t valueCoef.t valueCoef.t valueCoef.t value

Spatial autocorrelation coefficient0.102.670.09NANANANA

Coefficient for endogenous variable ()travel_time0.080.010.36NANANANA
pop_den0.270.35NANANANA

Coefficients for facility variables ()Constant−0.55−0.980.030.05−0.37−0.740.160.34
Office−0.50−0.39−0.42−0.36
Supermarket0.061.310.051.410.051.490.051.47
mid_school2.441.421.881.20
Clinic−0.59−0.44−0.51−0.50
Inn0.191.380.121.140.121.280.131.05
Sports−0.49−0.99−0.17−0.44−0.65−1.57−0.33−0.74

Coefficients for transportation variables ()Constant1.921.81NANANANA
road_density0.050.05NANANANA
bus_den0.020.460.020.53NANANANA

Covariance matrix ()1.25 (fixed)NA1.25 (fixed)NANANANANA
−0.70−1.58−0.98NANANANA
1.941.93NANANANA

Number of observations32

0.05 level; 0.1 level; NA: not applicable.