A Comparison of Bayesian and Sampling Theory Inferences in a Probit Model

HE PROBIT MODEL IS A POPULAR DEVICE for explaining binary choice decisions in econometrics. It has been used to describe choices such as labor force participation, travel mode, home ownership, and type of education. These and many more examples can be found in papers by Amemiya (1981) and Maddala (1983). Given the contribution of economics towards explaining such choices, and given the nature of data that are collected, prior information on the relationship between a choice probability and several explanatory variables frequently exists. Bayesian inference is a convenient vehicle for including such prior information. Given the increasing popularity of Bayesian inference it is useful to ask whether inferences from a probit model are sensitive to a choice between Bayesian and sampling theory techniques. Of interest is the sensitivity of inference on coefficients, probabilities, and elasticities. We consider these issues in a model designed to explain choice between fixed and variable interest rate mortgages. Two Bayesian priors are employed: a uniform prior on the coefficients, designed to be noninformative for the coefficients, and an inequality restricted prior on the signs of the coefficients. We often know, a priori, whether increasing the value of a particular explanatory variable will have a positive or negative effect on a choice probability. This knowledge can be captured by using a prior probability density function (pdf) that is truncated to be positive or negative. Thus, three sets of results are compared:those from maximum likelihood (ML) estimation, those from Bayesian estimation with an unrestricted uniform prior on the coefficients, and those from Bayesian estimation with a uniform prior truncated to accommodate inequality restrictions on the coefficients.

Testing for Market Power in Multiple-Input, Multiple-Output Industries: The Australian Grains and Oilseeds Industries

Recent empirical studies have found significant evidence of departures from competition in the input side of the Australian bread, breakfast cereal and margarine end-product markets. For example, Griffith (2000) found that firms in some parts of the processing and marketing sector exerted market power when purchasing grains and oilseeds from farmers. As noted at the time, this result accorded well with the views of previous regulatory authorities (p.358). In the mid-1990s, the Prices Surveillence Authority (PSA 1994) determined that the markets for products contained in the Breakfast Cereals and Cooking Oils and Fats indexes were "not effectively competitive" (p.14). The PSA consequently maintained price surveillence on the major firms in this product group. The Griffith result is also consistent with the large number of legal judgements against firms in this sector over the past decade for price fixing or other types of non-competitive behaviour. For example, bread manufacturer George Weston was fined twice during 2000 for non-competitive conduct and the ACCC has also recently pursued and won cases against retailer Safeway in grains and oilseeds product lines.

Estimating economic relationships subject to firm- and time-varying equality and inequality constraints

Applied econometricians often fail to impose economic regularity constraints in the exact form economic theory prescribes. We show how the Singular Value Decomposition (SVD) Theorem and Markov Chain Monte Carlo (MCMC) methods can be used to rigorously impose time- and firm-varying equality and inequality constraints. To illustrate the technique we estimate a system of translog input demand functions subject to all the constraints implied by economic theory, including observation-varying symmetry and concavity constraints. Results are presented in the form of characteristics of the estimated posterior distributions of functions of the parameters. Copyright (C) 2001 John Wiley Sons, Ltd.

The role of gender in reactions to service failure and recovery

Male and female consumers place different emphasis on elements of the service recovery process. Perceptions were influenced by gender of the service provider and by a match of customer and service provider gender. The study, an experimental design with 712 respondents, found that when service providers, irrespective of gender, display concern and give customers voice and a sizable compensation, both men and women reported more positive attitudes compared with when this was not so. Combinations of high voice with high outcome and high voice with high concern were especially important in positively influencing perceptions of effort, regardless of gender. However, the authors also found that there were significant differences between male and female respondents regarding their perceptions of how service recovery should be handled. Women want their views heard during service recovery attempts and to be allowed to provide input. Men, in contrast, do not view voice as important.

Aglomeração econômica e disparidade salarial no Espírito Santo : teoria e evidência a partir de um enfoque da Nova Geografia Econômica

O trabalho buscou analisar questões de desigualdade regional no Espírito Santo através da linha de pesquisa denominada Nova Geografia Econômica (NGE). Uma forma de realizar essa análise é através do estudo da relação entre diferenciais de salário e mercado potencial. Mais precisamente, o trabalho procurou verificar o impacto de fatores geográficos de segunda natureza – mercado potencial – nos salário médios municipais. Inicialmente, por meio de uma Análise Exploratória de Dados Espaciais, verificou-se que os salários são maiores próximos às regiões com alto mercado potencial (litoral/RMGV). Por meio da utilização de técnicas de estatística e econometria espacial foi possível observar para os anos de 2000 e 2010 a existência de uma estrutura espacial de salários no Espírito Santo. O coeficiente de erro autorregressivo foi positivo e estatisticamente significativo, indicando o modelo SEM (spatial error model) como o mais apropriado para modelar os efeitos espaciais. Os resultados indicam ainda que não só fatores educacionais afetam os salários, fatores geográficos de segunda natureza possuem um efeito até maior quando comparados aos primeiros. Conclui-se, como demonstra o modelo central da NGE que, forças exclusivamente de mercado nem sempre levam ao equilíbrio equalizador dos rendimentos, pelo contrário, levam à conformação de uma estrutura do tipo centro-periferia com diferença persistente de rendimentos entre as regiões. Adicionalmente, verifica-se que os municípios que apresentam maior salário, maior mercado potencial e melhores indicadores sociais são àqueles localizados no litoral do estado, mais precisamente os municípios próximos à RMGV. Sendo assim, o trabalho reforça a necessidade de que se pense estratégias que fomentem a criação de novas centralidades no Espírito Santo, a fim de atuar na redução das desigualdades regionais. O trabalho se insere num grupo de vários outros estudos que analisaram questões de desigualdade e concentração produtiva no Espírito Santo. A contribuição está na utilização do referencial teórico da NGE, que ainda não havia sido empregada para o estado, e na utilização de técnicas de estatística espacial e econometria espacial.

The Sticky Information Macro Model: Beyond Perfect Foresight

Sticky information monetary models have been used in the macroeconomic literature to explain some of the observed features regarding inflation dynamics. In this paper, we explore the consequences of relaxing the rational expectations assumption usually taken in this type of model; in particular, by considering expectations formed through adaptive learning, it is possible to arrive to results other than the trivial convergence to a fixed point long-term equilibrium. The results involve the possibility of endogenous cyclical motion (periodic and a-periodic), which emerges essentially in scenarios of hyperinflation. In low inflation settings, the introduction of learning implies a less severe impact of monetary shocks that, nevertheless, tend to last for additional time periods relative to the pure perfect foresight setup.

Accessibility and Local Development: Interaction between Cross-border Accessibility and Local Development in Portugal and Spain

One of de EU major concerns is cohesion and cross-border regional development. Usually cross-border regions are less dynamic, acting as bottlenecks mainly in peripheral territories. This paper is focused on the Portuguese-Spanish border using socio-economic and accessibility data. It considers Spatial Econometrics to produce statistical evidence on the relationship between accessibility and development at a local scale. A pilot study is conducted on North and Center region using variables such as population age, graduation characteristics, migrations, unemployment and daily accessibility to main towns in future this evaluation will be applied to the entire cross-border area between Portugal and Spain.

A fractional perspective to financial indices

The application of mathematical methods and computer algorithms in the analysis of economic and financial data series aims to give empirical descriptions of the hidden relations between many complex or unknown variables and systems. This strategy overcomes the requirement for building models based on a set of ‘fundamental laws’, which is the paradigm for studying phenomena usual in physics and engineering. In spite of this shortcut, the fact is that financial series demonstrate to be hard to tackle, involving complex memory effects and a apparently chaotic behaviour. Several measures for describing these objects were adopted by market agents, but, due to their simplicity, they are not capable to cope with the diversity and complexity embedded in the data. Therefore, it is important to propose new measures that, on one hand, are highly interpretable by standard personal but, on the other hand, are capable of capturing a significant part of the dynamical effects.

A estrutura conceptual da informação financeira e o governo das sociedades

A informação contabilística constitui um dos pilares fundamentais para a formação de juízos e tomada de decisões empresariais e, para isso, deve cumprir os requisitos de qualidade, objectividade, transparência, rigor, responsabilidade e independência. Para o efeito, a informação é elaborada com base em políticas e critérios contabilísticos, conceitos e pressupostos subjacentes à “Estrutura Conceptual” dos normativos emitidos por organismos nacionais e internacionais, e em princípios sustentados no “Governo das Sociedades” e respectivas recomendações, considerando as necessidades dos seus utilizadores. O presente estudo tem por objectivo explorar a relação entre a Informação Financeira Divulgada e o Modelo de Governação das Sociedades e fornecer uma melhor e mais profunda compreensão das características das sociedades que influenciam a divulgação. Neste estudo, foi utilizada uma amostra de empresas portuguesas não financeiras, emitentes de acções, que se encontram admitidas a negociação no mercado de cotações oficiais da New York Stock Exchange, adiante designada por NYSE Euronext Lisbon, no período de 2010 a 2012. Os elementos foram recolhidos através da leitura e análise dos Relatórios e Contas Anuais publicados e por consulta da base de dados DataStream. Os dados obtidos foram analisados através de modelos de regressão linear múltipla, considerando o modelo de dados em painel, tendo sido utilizado o programa de estatística Gnu Regression, Econometrics and Time- Series Library (GRETL), versão 1.9.92 (Sep 20, 2014), embora, para a análise das correlações, tenha sido utilizado, também, o programa estatístico Statistical Package for the Social Sciences, adiante denominado IBM SPSS Statistics, versão 19. A análise dos resultados indica que as principais determinantes subjacentes à divulgação relacionada com o índice de cumprimento sobre o Governo das Sociedades são as variáveis explicativas relacionadas com: (i) dimensão da empresa; (ii) endividamento; (iii) concentração accionista; (iv) negociabilidade; (v) percentagem de acções detidas pelos administradores executivos; (vi) rendibilidade do investimento total; e (vii) remuneração dos administradores não executivos. Os testes efectuados mostram que os resultados alcançados são robustos para modelar especificações e problemas de existência de multicolinearidade. O software utiliza o método de Arellano (2004) com a finalidade de corrigir o problema de possibilidade de existência de heteroscedasticidade e de autocorrelação. Nesta base, verificaram-se todas as situações que poderiam criar obstáculos e causar enviesamento nos estimadores e definições das variáveis. Relativamente ao índice global de divulgação dos riscos e incertezas, concluiu-se que é provavelmente determinado por diferentes variáveis. Esta investigação enriquece a discussão sobre as relações entre a divulgação da informação e a estrutura do Governo das Sociedades. Assim, o estudo pode ser útil para accionistas, administradores, credores e outros investidores, quanto à reflexão da adequação da informação divulgada nos relatórios e contas anuais, com a finalidade da tomada de decisões idóneas e fundamentadas. Porém, é essencial que os reguladores incentivem as empresas a adequarem melhor as práticas de governo e a criarem mais mecanismos de forma a alcançar um maior nível de independência e uma adopção mais ajustada das recomendações sobre o Governo das Sociedades.

Empiricism in ecological economics: a perspective from complex systems theory

Economies are open complex adaptive systems far from thermodynamic equilibrium, and neo-classical environmental economics seems not to be the best way to describe the behaviour of such systems. Standard econometric analysis (i.e. time series) takes a deterministic and predictive approach, which encourages the search for predictive policy to ‘correct’ environmental problems. Rather, it seems that, because of the characteristics of economic systems, an ex-post analysis is more appropriate, which describes the emergence of such systems’ properties, and which sees policy as a social steering mechanism. With this background, some of the recent empirical work published in the field of ecological economics that follows the approach defended here is presented. Finally, the conclusion is reached that a predictive use of econometrics (i.e. time series analysis) in ecological economics should be limited to cases in which uncertainty decreases, which is not the normal situation when analysing the evolution of economic systems. However, that does not mean we should not use empirical analysis. On the contrary, this is to be encouraged, but from a structural and ex-post point of view.

Sources of Competitiveness in Tourist Local Systems

At the end of the XIX Century, Marshall described the existence of some concentrations of small and medium enterprises specialised in a specific production activity in certain districts of some industrial English cities. Starting from his contribute, Italian scholars have paid particular attention to this local system of production coined by Marshall under the term industrial district. In other countries, different but related territorial models have played a central role as the milieu or the geographical industrial clusters. Recently, these models have been extended to non-industrial fields like culture, rural activities and tourism. In this text, we explore the extension of these territorial models to the study of tourist activities in Italy, using a framework that can be easily applied to other countries or regions. The paper is divided in five sections. In the first one, we propose a review of the territorial models applied to tourism industry. In the second part, we construct a tourist filiere and we apply a methodology for the identification of local systems through GIS tools. Thus, taxonomy of the Italian Tourist Local Systems is presented. In the third part, we discuss about the sources of competitiveness of these Tourist Local Systems. In the forth section, we test a spatial econometrics model regarding different kinds of Italian Tourist Local Systems (rural systems, arts cities, tourist districts) in order to measure external economies and territorial networks. Finally, conclusions and policy implications are exposed.