Abstract
This paper develops alternative stochastic frontier models (ASFM) for estimating time-space prism vertices with different distributional assumptions for the inefficiency term that takes a non-negative value. The traditional stochastic frontier model (SFM) assumes that the inefficiency term follows a half-normal or exponential distribution. Under those assumptions, most travelers’ home departure/arrival time will be close to prism vertices, which is not necessarily consistent with actual travel behaviors. To avoid this potential problem, the ASFM adopt alternative distributions for the inefficiency term whose density values can decrease monotonously or vary non-monotonously. Quasi-Monte Carlo simulation method is employed to estimate the ASFM without closed-form likelihood expressions. Simulation experiment results show that SFM needs a substantially greater number of Halton draws for consistent estimators than a typical mixed logit model does. The ASFM are estimated based on the travel data of 1454 Shanghai commuters and 2964 Houston commuters. It is found that models with inefficiency term following a half-normal distribution tend to underestimate the origin vertex of morning prism and overestimate the terminal vertex of evening prism over 50 and 30 min for Shanghai and Houston samples, respectively. The empirical results show the importance of choosing an appropriate distributional assumption for the inefficiency term in the SFM for better understanding the relation between individuals’ departure/arrival time and time-space prism vertices. The SFM based on an appropriate distributional assumption can be applied in activity-based models for big cities to better reflect tighter temporal constraints on metropolitan residents and narrower time-space prisms for outdoor activity arrangement.
Similar content being viewed by others
Notes
In this paper, when the inefficiency follows a half-normal, or exponential, or Rayleigh distribution, the likelihood functions of the SFMs can be expressed by the cumulative density function of standard normal distribution, which has numerical approximation and can be efficiently calculated in most statistical software. Thus, we consider that these three SFMs have closed-form likelihood functions in form of built-in functions. While the SFM with inefficiency log-normally distributed has a likelihood function with a one-dimensional integration needed to be evaluated. Therefore, when the inefficiency follows a log-normal distribution, we consider the SFM does not have a closed-form likelihood function.
References
Aigner, D., Lovell, C.K., Schmidt, P.: Formulation and estimation of stochastic frontier production function models. J. Econ. 6(1), 21–37 (1977)
Asgari, H., Jin, X., Rojas, M.B.: The impacts of telecommuting on the time-space distribution of daily activities. Paper presented at the 97th Annual Meeting of the Transportation Research Board, Wanshington, DC (2018)
Banerjee, A., Ye, X., Pendyala, R.M.: Understanding travel time expenditures around the world: exploring the notion of a travel time frontier. Transportation 34(1), 51–65 (2007)
Ben-Akiva, M., McFadden, D., Train, K., Walker, J., Bhat, C., Bierlaire, M., Bolduc, D., Boersch-Supan, A., Brownstone, D., Bunch, D.S.: Hybrid choice models: progress and challenges. Mark. Lett. 13(3), 163–175 (2002)
Bhat, C.R.: Quasi-random maximum simulated likelihood estimation of the mixed multinomial logit model. Transp. Res. Part B Methodol. 35(7), 677–693 (2001)
Bhat, C.R.: Simulation estimation of mixed discrete choice models using randomized and scrambled Halton sequences. Transp. Res. Part B Methodol. 37(9), 837–855 (2003)
Cox, D.R.: Tests of separate families of hypotheses. In: Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability, vol. 105, p. 23 (1961)
De Borger, B., Kerstens, K., Costa, A.: Public transit performance: what does one learn from frontier studies? Transp. Rev. 22(1), 1–38 (2002)
De Grange, L., Troncoso, R., Briones, I.: Cost, production and efficiency in local bus industry: an empirical analysis for the bus system of Santiago. Transp. Res. Part A Policy Pract. 108, 1–11 (2018)
Greene, W.H.: LIMDEP, Version 11.0. Econometric Modeling Guide, vol. 1. Econometric Software, Plainview, NY (2016)
Hägerstraand, T.: What about people in regional science? Pap. Reg. Sci. 24(1), 7–24 (1970)
Hajargasht, G.: Stochastic frontiers with a Rayleigh distribution. J. Product. Anal. 44(2), 199–208 (2015)
Hogg, R.V., McKean, J., Craig, A.T.: Introduction to Mathematical Statistics. Pearson Education, London (2005)
Holmgren, J.: The effects of using different output measures in efficiency analysis of public transport operations. Res. Transp. Bus. Manag. 28, 12–22 (2018)
Horowitz, J.L.: Statistical comparison of non-nested probabilistic discrete choice models. Transp. Sci. 17(3), 319–350 (1983)
Kitamura, R., Yamamoto, T., Kishizawa, K., Pendyala, R.M.: Stochastic frontier models of prism vertices. Transp. Res. Rec. J. Transp. Res. Board 1718(1), 18–26 (2000)
Kitamura, R., Yamamoto, T., Susilo, Y.O., Axhausen, K.W.: How routine is a routine? An analysis of the day-to-day variability in prism vertex location. Transp. Res. Part A Policy Pract. 40(3), 259–279 (2006)
Li, D., Miwa, T., Morikawa, T.: Analysis of vehicles’ daily fuel consumption frontiers with long-term controller area network data. Transp. Res. Rec. J. Transp. Res. Board 2503, 100–109 (2015)
Liao, F., Rasouli, S., Timmermans, H.: Incorporating activity-travel time uncertainty and stochastic space–time prisms in multistate supernetworks for activity-travel scheduling. Int. J. Geogr. Inf. Sci. 28(5), 928–945 (2014)
Meeusen, W., van Den Broeck, J.: Efficiency estimation from Cobb-Douglas production functions with composed error. Int. Econom. Rev. 18, 435–444 (1977)
Neutens, T., Schwanen, T., Witlox, F.: The prism of everyday life: towards a new research agenda for time geography. Transp. Rev. 31(1), 25–47 (2011)
Parmeter, C.F., Kumbhakar, S.C.: Efficiency analysis: a primer on recent advances. Found. Trends Econ. 7(3–4), 191–385 (2014)
Patil, P.N., Dubey, S.K., Pinjari, A.R., Cherchi, E., Daziano, R., Bhat, C.R.: Simulation evaluation of emerging estimation techniques for multinomial probit models. J. Choice Model. 23, 9–20 (2017)
Pendyala, R.M., Yamamoto, T., Kitamura, R.: On the formulation of time-space prisms to model constraints on personal activity-travel engagement. Transportation 29(1), 73–94 (2002)
Pinjari, A.R., Augustin, B., Sivaraman, V., Imani, A.F., Eluru, N., Pendyala, R.M.: Stochastic frontier estimation of budgets for Kuhn–Tucker demand systems: application to activity time-use analysis. Transp. Res. Part A Policy Pract. 88, 117–133 (2016)
Ravulaparthy, S.K., Konduri, K.C., Goulias, K.G.: Exploratory analysis of the activity time-use frontier and its effect on episodic well-being: data from the disability and use of time survey. Transp. Res. Rec. J. Transp. Res. Board. 2669, 80–90 (2017)
Scrogin, D., Hofler, R., Boyle, K., Walter Milon, J.: An efficiency approach to choice set formation: theory and application to recreational destination choice. Appl. Econ. 42(3), 333–350 (2010)
Soo, J.: Space-time prism vertices: exploring gender differences and multiple-peak distributions in arrival and departure times. Eur. J. Trans. Infrastruct. Res. 9(4), 380–396 (2009)
Sun, X.-H., Yamamoto, T., Morikawa, T.: Stochastic frontier analysis of excess access to mid-trip battery electric vehicle fast charging. Transp. Res. Part D Transp. Environ. 34, 83–94 (2015)
Susilo, Y.O., Avineri, E.: The impacts of household structure on the individual stochastic travel and out-of-home activity time budgets. J. Adv. Transp. 48(5), 454–470 (2014)
Timmermans, H., Arentze, T., Joh, C.-H.: Analysing space-time behaviour: new approaches to old problems. Prog. Hum. Geogr. 26(2), 175–190 (2002)
Tosa, C., Miwa, T., Morikawa, T.: Space-time prism vertices for morning and evening commuters in Cluj-Napoca, Romania. In: Advanced Engineering Forum, pp. 573–578. Trans Tech Publications (2017)
Train, K.E.: Discrete Choice Methods with Simulation. Cambridge University Press, Cambridge (2009)
Wang, K., Ye, X., Pendyala, R.M., Zou, Y.: On the development of a semi-nonparametric generalized multinomial logit model for travel-related choices. PLoS ONE 12(10), e0186689 (2017)
Weiss, A., Habib, K.N.: Who is picking up the kid from daycare? Understanding the intra-household dynamics in drop-off and pick-up allocation for households with dependent children. Transp. Res. Rec. J. Transp. Res. Board. 2668, 51–59 (2017)
Yamamoto, T., Kitamura, R., Pendyala, R.M.: Comparative analysis of time-space prism vertices for out-of-home activity engagement on working and nonworking days. Environ. Plan. B Plan. Des. 31(2), 235–250 (2004)
Ye, X., Pendyala, R.M.: Understanding time-space prism constraints for modeling weekend travel behavior. In: Traffic and Transportation Studies-Proceedings of the Fourth International Conference on Traffic and Transportation Studies (2004)
Ye, X., Wang, K., Zou, Y., Lord, D.: A semi-nonparametric Poisson regression model for analyzing motor vehicle crash data. PLoS ONE 13(5), e0197338 (2018)
Acknowledgements
This research is partially supported by the general project “Study on the Mechanism of Travel Pattern Reconstruction in Mobile Internet Environment” (No. 71671129) and the key project “Research on the Theories for Modernization of Urban Transport Governance” (No. 71734004) from the National Natural Science Foundation of China.
Author information
Authors and Affiliations
Contributions
KW: Literature review, data collection and analysis, manuscript writing. XY: conceptual Idea, study planning, manuscript writing and editing.
Corresponding author
Ethics declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
Derivation of Eq. (5)
Since \( g(u) = \frac{2}{{\sigma_{u} \sqrt {2\pi } }}{\kern 1pt} {\kern 1pt} {\kern 1pt} e^{{ - \frac{{u^{2} }}{{2\sigma_{u}^{2} }}}} ,{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} \sigma_{u} > 0 \); \( h(v) = \frac{1}{{\sigma_{v} \sqrt {2\pi } }}{\kern 1pt} {\kern 1pt} {\kern 1pt} e^{{ - \frac{{v^{2} }}{{2\sigma_{v}^{2} }}}} \), \( \varepsilon = v - Su \);
Let \( \sigma^{2} = \sigma_{u}^{2} + \sigma_{v}^{2} \), \( \lambda = \sigma_{u} /\sigma_{v} \);
Define \( \frac{\sigma }{{\sigma_{u} \sigma_{v} }}\left( {u + \frac{{S\sigma_{u}^{2} \varepsilon }}{{\sigma^{2} }}} \right) = t \), then \( \frac{\sigma }{{\sigma_{u} \sigma_{v} }}du = dt \); \( du = \frac{{\sigma_{u} \sigma_{v} }}{\sigma }dt \).
When \( u = 0 \), \( t = \frac{\sigma }{{\sigma_{u} \sigma_{v} }} \cdot \frac{{S\sigma_{u}^{2} \varepsilon }}{{\sigma^{2} }} = \frac{S\varepsilon \lambda }{\sigma } \); when \( u \to + \infty \), \( t \to + \infty \). Then,
Derivation of Eq. (6)
Since \( g(u) = \frac{1}{{\sigma_{u} }}{\kern 1pt} {\kern 1pt} {\kern 1pt} e^{{ - \frac{u}{{\sigma_{u} }}}} ,{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} \sigma_{u} > 0 \), \( h(v) = \frac{1}{{\sigma_{v} \sqrt {2\pi } }}{\kern 1pt} {\kern 1pt} {\kern 1pt} e^{{ - \frac{{v^{2} }}{{2\sigma_{v}^{2} }}}} \), \( \varepsilon = v - Su \);
Let \( \frac{u}{{\sigma_{v} }} + \left( {\frac{{\sigma_{v} }}{{\sigma_{u} }} + \frac{\varepsilon S}{{\sigma_{v} }}} \right) = t \), then \( \frac{1}{{\sigma_{v} }}du = dt \);\( du = \sigma_{v} dt \).
When \( u = 0 \), \( t = \frac{{\sigma_{v} }}{{\sigma_{u} }} + \frac{\varepsilon S}{{\sigma_{v} }} \); when \( u \to + \infty \), \( t \to + \infty \). Then,
Derivation of Eq. (7)
Since \( g(u) = \frac{u}{{\sigma_{u}^{2} }} \cdot \exp \left( { - \frac{{u^{2} }}{{2\sigma_{u}^{2} }}} \right),{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} \sigma_{u} > 0 \), \( h(v) = \frac{1}{{\sigma_{v} \sqrt {2\pi } }}{\kern 1pt} {\kern 1pt} {\kern 1pt} e^{{ - \frac{{v^{2} }}{{2\sigma_{v}^{2} }}}} \), \( \varepsilon = v - Su \);
Let \( \mu = \frac{{ - \varepsilon S\sigma_{u}^{2} }}{{\sigma_{u}^{2} + \sigma_{v}^{2} }} \), \( \sigma_{r}^{2} = \frac{{\sigma_{u}^{2} \sigma_{v}^{2} }}{{\sigma_{u}^{2} + \sigma_{v}^{2} }} \); then,
Let \( t = \frac{u - \mu }{{\sigma_{r} }} \), then \( u = t\sigma_{r} + \mu \); \( dt = \frac{{1{\kern 1pt} {\kern 1pt} }}{{\sigma_{r} }}du \).
Let \( u = \frac{{t^{2} }}{2} \), \( du = t{\kern 1pt} {\kern 1pt} dt \).
When \( t = - \frac{\mu }{{\sigma_{r} }} \), \( u = \frac{{\mu^{2} }}{{2\sigma_{r}^{2} }} \); when \( u \to + \infty \), \( t \to + \infty \). Then,
Derivation of Eq. (8)
Since \( g(u) = \frac{1}{{u\sigma_{u} }} \cdot \varphi \left( { - \frac{{\ln (u) - u_{l} }}{{\sigma_{u} }}} \right),\sigma_{u} > 0,{\kern 1pt} {\kern 1pt} {\kern 1pt} - \infty < u_{l} < \infty \), \( h(v) = \frac{1}{{\sigma_{v} \sqrt {2\pi } }}{\kern 1pt} {\kern 1pt} {\kern 1pt} e^{{ - \frac{{v^{2} }}{{2\sigma_{v}^{2} }}}} \), \( \varepsilon = v - Su \);
Rights and permissions
About this article
Cite this article
Wang, K., Ye, X. Development of alternative stochastic frontier models for estimating time-space prism vertices. Transportation 48, 773–807 (2021). https://doi.org/10.1007/s11116-019-10056-0
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11116-019-10056-0