|
on Discrete Choice Models |
| By: | Pene Kalulumia (Département d'économique, Université de Sherbrooke); Denis Bolduc (GREEN Département d'économique Université Laval, Québec G1K 7P4) |
| Abstract: | In this paper, we applied the generalized mixed estimation approach to the problem of estimating the Quebec residential electricity demand for space and water heating. A multinomial discrete-continuous choice model is used and estimated in two stages. The discrete choice is modelled as a multinomial probit model, while the continuous choice is estimated from a reduced form approach which corrects for the simultaneity biases. The results indicate that the GM estimator which combines prior and sample information dominates the classical ML estimator of the MNP models and hence, provides better prevision for electricity consumption. Evidence also shows that heating-system capital and operating costs, households characteristics, and energy prices have a significant impact on the choice of heating systems and electricity use. In particular, price substitution effects are well predicted. |
| Keywords: | Generalized mixed estimator, residential electricity demand, multinomial probit, discretecontinuous choice. |
| JEL: | C C13 C51 D12 Q41 |
| Date: | 2004 |
| URL: | http://d.repec.org/n?u=RePEc:shr:wpaper:04-01&r=dcm |
| By: | Isabel Proenca (ISEG-UTL); Joao Santos Silva (ISEG-UTL) |
| Abstract: | Although semiparametric alternatives are available, parametric binary choice models are widely used in practice, in spite of their sensitivity to misspecification. Here we present the results of a simulation study on the finite sample performance of parametric and semiparametric specification tests for this kind of models. The results obtained indicate that the computationally demanding semiparametric tests do not generally outperform the simpler score tests against parametric alternatives. |
| JEL: | C12 C14 C25 C52 |
| Date: | 2005–08–05 |
| URL: | http://d.repec.org/n?u=RePEc:wpa:wuwpem:0508008&r=dcm |
| By: | Walter Beckert (Institute for Fiscal Studies and Birkbeck College London); Richard Blundell (Institute for Fiscal Studies and University College London) |
| Abstract: | This paper considers structural nonparametric random utility models for continuous choice variables. It provides suffcient conditions on random preferences to yield reduced- form systems of nonparametric stochastic demand functions that allow global invertibility between demands and random utility components. Invertibility is essential for global identification of structural consumer demand models, for the existence of well-specified probability models of choice and for the nonparametric analysis of revealed stochastic preference. |
| Keywords: | nonparametric random utility model, stochastic demand, global invertibility |
| JEL: | C14 C31 C51 D1 |
| Date: | 2005–07 |
| URL: | http://d.repec.org/n?u=RePEc:ifs:cemmap:09/05&r=dcm |
| By: | Carlos Barros (ISEG-UTL); Isabel Proenca (ISEG-UTL); Jose Cabral Vieira (University of Azores) |
| Abstract: | In this paper, we examine the determinants of low-wage employment in Portugal. For this purpose, we use a data file of the European Community Household Panel (ECHP) for the years 1998 and 1999. In order to take into account unobserved heterogeneity in the data, a random-parameter logit model is used to analyse the probability of a worker receiving a low wage. The results indicate that the consideration that the effects of the explanatory variables are the same across all individuals, such as is assumed in most of the literature may be misleading. From the policy perspective, this implies that the use of a single instrument in order to combat low-wage employment is inappropriate to satisfy the whole population. In view of this, policies tailored by clusters would be more appropriate. |
| Keywords: | low-wage employment, random-parameter logit model, public policy. |
| JEL: | C25 J31 J38 |
| Date: | 2005–08–05 |
| URL: | http://d.repec.org/n?u=RePEc:wpa:wuwpla:0508001&r=dcm |
| By: | Reto Nyffeler |
| Abstract: | In recent applications of discrete choice models of labour supply considerable attention has been devoted to strategies to increase the flexibility of models for a better fit to the data. These include the introduction of random parameters, fixed cost of work or flexible functional forms of preferences. Based on estimates of models of recent studies this paper compares these different modeling strategies. Results for Swiss data show that the traditional way to interpret fixed cost of work is ad hoc. Furthermore our results indicate that care should be taken when using very general function forms of preferences |
| Keywords: | multinomial logit; household labour supply; taxation; microsimulation |
| JEL: | C25 C52 H31 J22 |
| Date: | 2005–07 |
| URL: | http://d.repec.org/n?u=RePEc:ube:dpvwib:dp0508&r=dcm |
| By: | Melanie K. Jones (WELMERC, University of Wales Swansea); Richard J. Jones (WELMERC, University of Wales Swansea); Philip D. Murphy (WELMERC, University of Wales Swansea); Peter J. Sloane (WELMERC, University of Wales Swansea and IZA Bonn) |
| Abstract: | An important policy issue is whether the National Minimum Wage (NMW) introduced in Britain in April 1999, is a stepping stone to higher wages or traps workers in a low-wage – nowage cycle. In this paper we utilise the longitudinal element of the Labour Force Survey over the period 1999 to 2003 to model transitions between different labour market states – payment at or below the NMW, above the NMW, unemployment and inactivity, using a multinomial logit approach. It appears that for many workers payment at or below the NMW is of relatively short duration and a substantial number move into higher paid jobs. |
| Keywords: | national minimum wage, transitions, steady state distributions |
| JEL: | J0 J3 J6 |
| Date: | 2005–07 |
| URL: | http://d.repec.org/n?u=RePEc:iza:izadps:dp1690&r=dcm |
| By: | François Ortalo-Magné; Sven Rady |
| Abstract: | Standard explanations for the income heterogeneity within neighborhoods rely on differences of preferences across households and heterogeneity of the housing stock. We propose an alternative and complementary explanation. We construct a stochastic equilibrium sorting model where (1) income is the sole dimension of household heterogeneity, (2) households form state-contingent housing location plans that may involve moves over their lifetimes, (3) households choose whether to own or rent depending on the housing expenditure risk associated with each tenure mode, and (4) there is a probability that newcomer households move in and compete for homes with native households. Income mixing within neighborhood arises for two reasons. First, allowing natives to form state-contingent housing location plans breaks the indivisibility of housing consumption implicit in the literature where households choose their location once and for all. Second, natives can insure themselves against rent fluctuations by buying their home prior to the realization of the population shock; newcomers cannot. As a result, poorer natives stay in the more desirable communities and only richer newcomers move in these communities. Evidence from U.S. metropolitan areas supports the effects predicted by the model. |
| JEL: | D31 R12 R21 |
| Date: | 2005 |
| URL: | http://d.repec.org/n?u=RePEc:ces:ceswps:_1465&r=dcm |
| By: | Jürgen Maurer (Institute for Fiscal Studies); André Meier |
| Abstract: | Recent theoretical contributions have suggested consumption externalities, or peergroup effects, as a potential explanation for some of the puzzles in macroeconomics and finance. However, the empirical relevance of peer effects for intertemporal consumption choice is a completely open question. To shed some light on the issue, we derive an extension of the standard life-cycle model that allows for consumption externalities. The analysis is complicated by the challenge of disentangling actual peer effects from merely correlated effects operating through common features or shocks within peer groups. We show how to conduct reliable inference under these circumstances based on within-group equilibrium conditions that give rise to a social multiplier. This approach can be understood as an adaptation of Manski’s "reflection problem framework" to the case of dynamic models with endogenous regressors. We estimate our model using US panel data from the PSID. While there is strong predictable consumption co-movement within peer groups, the evidence for true consumption externalities vanishes once correlated effects are adequately accounted for. |
| Keywords: | Consumption, Life-Cycle Model, Peer Effects, Reflection Problem |
| JEL: | C23 D12 D91 Z13 |
| Date: | 2005–07 |
| URL: | http://d.repec.org/n?u=RePEc:ifs:ifsewp:05/15&r=dcm |
| By: | Elena Argentesi |
| Abstract: | This paper looks at a form of non-price competition that has taken place in the Italian newspaper market, whereby weekly supplements are sold with the newspaper at a higher price. I estimate the impact of this selling strategy using a logit and a nested logit model of demand on a panel of Italian newspapers. I show that supplements increase the readership both in the weekday of issue and in the average weekday. This suggests that supplements are a way to attract new readers for the newspaper. This promotional effect is due both to business stealing and to market expansion. |
| JEL: | L11 L82 C33 |
| Date: | 2004 |
| URL: | http://d.repec.org/n?u=RePEc:eui:euiwps:eco2004/28&r=dcm |