TABLE 2 Estimated Discrete Choice Functions,,,,,
,MNL,NMNL,,,
,,Tree1,Tree2,Tree3,Tree4
Performance statistics,,,,,
Log likelihood,"–3,326.46","–3,326.10","–3,272.11","–3,325.52","–3,325.71"
McFadden Rho-squared,0.4040,0.4041,0.4038,0.4042,0.4042
Correctly predicted (%),58.746,58.746,58.705,58.746,58.746
Alternative specific variables,,,,,
Distance,–0.37(–37.0),–0.80(–40.3),–0.37(–36.8),–0.35(–16.9),–0.38(–31.8)
TEUs per berth,–2.05(–4.5),–4.37(–9.4),–1.55(–3.3),–1.95(–4.2),–2.09(–4.6)
TEUs per crane,4.01(6.5),8.57(13.1),3.70(5.5),3.80(5.8),4.08(6.5)
Ship calls,–0.91(–7.0),–1.95(–14.0),–0.72(–5.0),–0.86(–6.2),–0.93(–7.1)
Usage factor,–3.27(–3.7),–7.09(–8.0),–2.82(–3.2),–3.09(–3.5),–3.33(–3.8)
Routes,0.10(6.6),0.22(13.2),0.07(4.1),0.10(5.9),0.11(6.6)
Constant variables,,,,,
Chinese,1.13(23.9),1.07(13.3),11.33(12.8),1.19(17.3),1.18(17.7)
"Dalian, Qingdao, other ports",0.74(0.8),1.50(1.5),0.69(0.7),0.71(0.7),0.76(0.8)
"Tianjin, Qingdao",–1.41(–1.4),–3.15(–3.2),–1.23(–1.3),–1.33(–1.4),–1.44(–1.5)
"Shanghai, other ports",3.75(4.4),7.91(9.0),3.30(3.7),3.55(4.1),3.81(4.4)
Dissimilarity parameters,,,,,
Chinese,1.00,2.16(49.7),,0.93(1.2),1.00
Non-Chinese,1.00,2.08(18.2),,1.00,1.06(1.2)
Dalian,,,9.10(7.6),,
Tianjin,,,12.20(7.0),,
Qingdao,,,3.82(7.0),,
Shanghai,,,12.7(30.0),,
Other ports,,,12.56(7.6),,
Note: Numbers in parentheses are t-statistics.,,,,,