Skip to main content

An Open Access Journal

Table 2 Attribute levels considered and prior parameters (pilot and final surveys) as input of Ngene software

From: Stated preference survey for estimating passenger transfer penalties: design and application to Madrid

Utility function

Coefficients

Prior parameters

Attributes

Levels

Pilot survey

Final survey

U (T0): no-transfer

tveh00

−.2000

−.3573

tveh0

35, 40, 45

mode00

1.0000

.5000*

mode0

0, 1

U (T1): one transfer

constant1

−1.0000

−3.9160

constant1

tveh01

−.2000

−.3705

tveh0

9, 12, 15

mode01

2.0000

.6236

mode0

0, 1

twalk11

−.2500

−.3407

twalk1

2, 4, 6

twait11

−.2000

−.3320

twait1

2, 5, 8

stair11

−1.0000

−.7000*

stair1

0, 1

tveh11

−.2000

−.3448

tveh1

9, 12, 15

mode11

2.0000

.6000*

mode1

0, 1

info1

2.0000

.7000*

info

0, 1

crowd1

−2.0000

−1.0272

crowd

0, 1

U (T2): two transfers

constant2

−1.1000

−5.7962

constant2

tveh02

−.2000

−.3255

tveh0

4, 7, 10

mode02

1.1000

1.5000*

mode0

0, 1

twalk12

−.4000

−.3570

twalk1

2, 3, 4

twait12

−.3000

−.4379

twait1

1, 3, 5

stair12

−1.0000

−1.1385

stair1

0, 1

tveh12

−.2000

−.4372

tveh1

4, 7, 10

mode12

1.1000

1.5490

mode1

0, 1

twalk22

−.4000

−.9148

twalk2

2, 3, 4

twait22

−.3000

−.3712

twait2

1, 3, 5

stair22

−1.0000

−.7091

stair2

0, 1

tveh22

−.2000

−.2790

tveh2

4, 7, 10

mode22

1.1000

1.5000*

mode2

0, 1

info2

1.1000

.7000*

info

0, 1

crowd2

−1.1000

−2.3642

crowd

0, 1

  1. *Prior parameters differ from the results of the MNL model as the parameters were not significantly different from zero at the 95% level