China
Africa
Explore chapters and articles related to this topic
Hidden Markov Models in Time Series, with Applications in Economics
Published in Sylvia Frühwirth-Schnatter, Gilles Celeux, Christian P. Robert, Handbook of Mixture Analysis, 2019
A few remarks are in order. First, in both functional forms, we do not fix the reference state to g0 = 1 as is usually done. In estimation, this generalization allows us to apply the permutation sampler to the reference state as well. Second, for G = 2, the probit specification has the advantage that a corresponding random utility model exists for g ≠ g0 with a latent variable utg* following a normal distribution: () utg*=x2tγzt−1,gx+γzt−1,g+νtg,νtg~N(0,1).
Applied Methods of Valuation of Water-related Ecosystem Services
Published in Robert A. Young, John B. Loomis, Determining the Economic Value of Water, 2014
Robert A. Young, John B. Loomis
The primary distinction between CM and CVM is in how respondents are asked about their WTP. In contrast to a CVM where a WTP question is asked for a single “with action” program or policy, CM presents the respondent with a set of alternative programs or management actions, each characterized by multiple attributes or characteristics of a particular program. Then each respondent is typically asked to choose their most preferred alternative. Each choice set has a “no change/current condition/status quo” alternative usually adjacent to one or more proposed action alternative. The alternative chosen is viewed, in a random utility model sense, to be the one with the highest net utility. Because one of the attributes included in each alternative is a price or costs as an attribute of each of the choice sets, the utility of each attribute can be monetized.
Smart Energy Resources: Supply and Demand
Published in Stuart Borlase, Smart Grids, 2018
Stuart Borlase, Sahand Behboodi, Thomas H. Bradley, Miguel Brandao, David Chassin, Johan Enslin, Christopher McCarthy, Stuart Borlase, Thomas Bradley, David P. Chassin, Johan Enslin, Gale Horst, Régis Hourdouillie, Salman Mohagheghi, Casey Quinn, Julio Romero Aguero, Aleksandar Vukojevic, Bartosz Wojszczyk, Eric Woychik, Alex Zheng, Daniel Zimmerle
From a utility planning and operation perspective, predicting the demand curve for RTP presents an additional challenge, particularly in response to a price disturbance after which the demand response resource may be depleted. From a first-principles approach, there is evidence supported by field demonstrations using transactive control systems [47,48] that the random utility model [49] best describes the demand curves of thermostatically controlled loads that respond primarily to short-term real-time price fluctuations [50].
Examining the effect of integrated ticketing on mode choice for interregional commuting: Studies among car commuters
Published in International Journal of Sustainable Transportation, 2023
I. B. Alhassan, B. Matthews, J. P. Toner, Y. O. Susilo
A binary logit (BNL) and binary mixed logit (ML) were specified and estimated using the cross-sectional dataset presented in the preceding section. The logit family of models provides a useful toolkit for analyzing and understanding discrete choice behaviors. The standard logit model, even though easy to estimate, does not consider random taste variation and has restricted substitution pattern (Train, 2009). The mixed logit model solves these problems. This model currently represents the state-of-the-art method in discrete choice modeling since it is a very flexible model that approximate any random utility model (Ortúzar et al., 2011; Train, 2009). For a detail description of the mathematical formulations and applications of these methods, the reader is referred to (Ben-Akiva & Lerman, 1985; Hensher & Greene, 2003; Koppelman & Bhat, 2006; Ortúzar et al., 2011; Train, 2009).
How does interchange affect passengers’ route choices in urban rail transit? – a case study of the Shanghai Metro
Published in Transportation Letters, 2022
Yan Cheng, Xiafei Ye, Taku Fujiyama
No matter how interchange is quantified at the route level, its influence on route choice is usually investigated using discrete choice models based on random utility theory (Ben-Akiva, Lerman, and Lerman 1985). The multinomial logit model (MNL) is the most popular model because of its closed-form formula and ease of interpretation, however, the independence from irrelevant alternatives (IIA) property derived from the assumption of independent distributions largely limits the model’s applicability and fidelity. Nested logit and other variations were then developed to overcome these problems, followed by great progress in the new Generalized Extreme Value (GEV) family of models and mixed logit models (McFadden 2001), which provide more flexibility. Particularly, mixed logit models can approximate any random utility model (McFadden and Train 2000). A more advanced version, the hybrid choice model has now been developed based on the discrete choice model to enhance the predictability of choice models (Ben-Akiva et al. 2002).
Radiation knowledge and willingness to buy bottled water from regions near the Fukushima Daiichi Nuclear Power Plant
Published in Water International, 2020
Our choice model is drawn from Lancaster (1966), showing that consumer utility is derived more by the characteristics or attributes that the goods possess than the goods themselves. Since McFadden (1974), application of such choice models with a random utility model (RUM) became very popular because while it is unrealistic for an empirical researcher to create a complete consumer utility function by knowing consumer’s complete attributes, RUM can capture the unobserved attributes in its nondeterministic component (Manski, 1977). Hence, the consumer choice behaviour in this study is also analysed with a random utility model. The model is estimated with a probit model. Defining as the utility of individual i when choosing choice k from K alternatives, a random utility is generally written as: