Measuring Travel Time Values with a Discrete Choice Model: A Note
Measuring Travel Time Values with a Discrete Choice Model: A Note
复制标题
使用离散选择模型测量行程时间值：注释
DOI：
10.2307/2232894
复制
发表时间：
1987
期刊：
影响因子：
--
通讯作者：
J. Bates
中科院分区：
文献类型：
--
作者：
J. Bates
在这本杂志最近的一篇文章中，Truong和Hensher(I985)开始提供有用的服务，将时间分配理论与离散选择模型的框架联系起来，从而允许在更合理的理论基础上实际估计时间的价值。不幸的是，他们的工作中有少数严重的误解，这导致混乱，特别是使他们从他们提出的经验性工作中得出的结论无效。本文旨在纠正Truong和Hensher论文中的理论错误，并根据这些更正，从他们提供的经验证据中得出一些简短的结论。由于他们所说的大多数都是成立的，所以这里只重复基本要点，并将在他们的论文中大量引用这些方程(方程式编号前用TH表示)。作者首先概述了Becker(I965)和DeSerpa(I97I)的开创性工作及时估值。这两种理论都依赖于效用最大化。由于DeSerpa的方法可以被视为Becker方法的扩展，我们将列出其要点，如作者的论文所给出的那样。对于手头的问题，我们假设个人有直接的效用函数，它依赖于G，他们可以购买的商品和服务的数量，L，他们有多少空闲时间，以及T他们必须花在旅行上的时间。旅行可以采取不同的方式，涉及不同的费用和旅行时间。由于总的金钱和时间预算是固定的，这些旅行成本和时间会影响其他商品的数量和可用的空闲时间。问题可以表述如下：Max u(Gi，Li，Ti)(Th i 2‘)使得G_1和L_t；M-C，(T_I_3)
In a recent article in this JOURNAL, Truong and Hensher (I985) set out to perform a useful service by relating the theory of time allocation to the framework of discrete choice models, thereby allowing practical estimation of the value of time on a sounder theoretical basis. Unfortunately, there are a small number of crucial misunderstandings in their work, which lead to confusion and, in particular, invalidate the conclusions they draw from their presented empirical work. This note aims to correct the errors of theory in Truong and Hensher's paper, and to draw some brief conclusions from the empirical evidence that they give, in the light of these corrections. Since most of what they say can stand, only the essential points are repeated here, and considerable reference will be made to the equations in their paper (indicated by TH before an equation no.). The authors begin by outlining the seminal work in time valuation of Becker (I965) and DeSerpa (I97I). Both theories rely on utility maximisation. Since DeSerpa's approach can be viewed as an extension of Becker's, we will set out its essentials, as given in the authors' paper. For the purpose of the problem in hand, we assume that individuals have direct utility functions which are dependent on G, the volume of goods and services they can buy, L, the amount of 'leisure' time they have, and T the amount of time they have to spend travelling. Travel can be by different modes i, involving different costs and travelling time. Since total money and time budgets are fixed, these travel costs and times impact on the amount of other goods and the amount of leisure time available. The problem can be set out as follows: Max u (Gi, Li, Ti) (TH I 2') such that G1 < M-C , (TH I3)