970 resultados para multinomial logit model
Resumo:
This paper traces the developments of credit risk modeling in the past 10 years. Our work can be divided into two parts: selecting articles and summarizing results. On the one hand, by constructing an ordered logit model on historical Journal of Economic Literature (JEL) codes of articles about credit risk modeling, we sort out articles which are the most related to our topic. The result indicates that the JEL codes have become the standard to classify researches in credit risk modeling. On the other hand, comparing with the classical review Altman and Saunders(1998), we observe some important changes of research methods of credit risk. The main finding is that current focuses on credit risk modeling have moved from static individual-level models to dynamic portfolio models.
Resumo:
This study explores the issue of internationalization through forming alliances with foreign capital in the small business sector in Turkey. Using a sample of 257 SMEs from this emerging market economy, collected via a field study, it finds that Turkish SMEs would like to form alliances with foreign capital for expanding their production capacity and accessing to world markets. Multivariate multinomial logit models are employed to analyse the survey data econometrically. Size-specific, sector-specific and management-specific factors are identified in the alliance motivation. The `market' and `finance' aspects of alliances prevail in the multivariate analysis with significant implications. There is also evidence that some conditional relationships offered by multivariate analysis differ from unconditional associations found in the survey, which implies that alliance motivation is partly a product of multidimensional decision-making process on behalf of SMEs.
Resumo:
The aim of this paper is to analyse the influence of a company's level of earnings and growth opportunities in determining the dividend policy choice of Malaysian-listed firms. The analysis is based on a sample of 136 firms listed on the Bursa Malaysia Index over a period of six years, from 1990 to 1996. The evidence suggests that the payers are more profitable than non-payers. Likewise, investment opportunity, which is measured by (∂At /At-1) and (Vt /At), differed for both payers and non-payers. The regression estimates from Logit model suggest that the average coefficient for EATA is a significant determinant for firm's dividend policy choice in Malaysia. This is consistent with the supposition that profitable firms are more likely to pay dividends than less profitable firms. Although investment opportunities, the firm's size and leverage were not found to be statistically significant, they provided some explanation for the dividend policy choice.
Resumo:
Although research on international travelers abounds in the literature, international travel by students remains a neglected area of research. This article reports the findings from a survey of 370 university students in New Zealand. The survey identified student motives for undertaking international travel, the planning process, and the preferred destinations and methods of financing international trips. Logit models were developed and estimated for two important aspects of international travel by students. In addition, the study also included cross-cultural comparisons of travel behavior. The findings indicate that students traveling overseas represent a distinct market with specific needs and preferences. Travel behaviors vary significantly for different cultures and the article shows that it is possible to model such behaviors.
Resumo:
The development and application of computational data mining techniques in financial fraud detection and business failure prediction has become a popular cross-disciplinary research area in recent times involving financial economists, forensic accountants and computational modellers. Some of the computational techniques popularly used in the context of - financial fraud detection and business failure prediction can also be effectively applied in the detection of fraudulent insurance claims and therefore, can be of immense practical value to the insurance industry. We provide a comparative analysis of prediction performance of a battery of data mining techniques using real-life automotive insurance fraud data. While the data we have used in our paper is US-based, the computational techniques we have tested can be adapted and generally applied to detect similar insurance frauds in other countries as well where an organized automotive insurance industry exists.
Resumo:
This study reports the findings of a choice experiment designed to explore local population preferences toward wetland ecosystem restoration of Bung Khong Long Wetland in Thailand. By addressing ecological, socioeconomic and cultural dimensions of ecosystem services, the findings provide policy-makers with a richer insight into the interconnections among ecological, socioeconomic and cultural systems in explaining the value of ecosystem services. Gaining an understanding of the trade-offs associated with different interests in ecosystem uses in this community has the capacity to promote wetland management and enhance land use planning. The choice experiment application entails selecting attributes and their levels and developing an experimental design to create the choice sets or hypothetical scenarios for welfare assessment via the questionnaire. The study is based on household level data collected from 780 randomly drawn respondents living around the lake and the data are analysed using the Random Parameter Logit Model with interactions. The findings indicate that the local population derives positive and significant values from the restoration of wetland ecosystem services, indicating caution is needed in the decision-making processes involving sensitive environments faced with competing uses. Socioeconomic and attitudinal characteristics of respondents are important factors influencing willingness to pay, implying community preferences are important in the effectiveness of environmental conservation efforts in this community. The cultural values associated with the wetland are significant suggesting that incorporating culture preferences may be a key factor in supporting wetland conservation.
Resumo:
INTRODUCTION: Nursing bedside handover in hospital has been identified as an opportunity to involve patients and promote patient-centred care. It is important to consider the preferences of both patients and nurses when implementing bedside handover to maximise the successful uptake of this policy. We outline a study which aims to (1) identify, compare and contrast the preferences for various aspects of handover common to nurses and patients while accounting for other factors, such as the time constraints of nurses that may influence these preferences.; (2) identify opportunities for nurses to better involve patients in bedside handover and (3) identify patient and nurse preferences that may challenge the full implementation of bedside handover in the acute medical setting. METHODS AND ANALYSIS: We outline the protocol for a discrete choice experiment (DCE) which uses a survey design common to both patients and nurses. We describe the qualitative and pilot work undertaken to design the DCE. We use a D-efficient design which is informed by prior coefficients collected during the pilot phase. We also discuss the face-to-face administration of this survey in a population of acutely unwell, hospitalised patients and describe how data collection challenges have been informed by our pilot phase. Mixed multinomial logit regression analysis will be used to estimate the final results. ETHICS AND DISSEMINATION: This study has been approved by a university ethics committee as well as two participating hospital ethics committees. Results will be used within a knowledge translation framework to inform any strategies that can be used by nursing staff to improve the uptake of bedside handover. Results will also be disseminated via peer-reviewed journal articles and will be presented at national and international conferences.
Resumo:
O objetivo geral desta tese é estimar a elasticidade-preço da demanda de forma conjunta, decomposta em elasticidade-preço da escolha da marca e da quantidade comprada, e discutir as implicações desta decomposição para uma categoria específica de produto. Para isto foram usados dados escaneados de uma amostra de domicílios no contexto varejista brasileiro. Oito hipóteses foram testadas por meio de dois modelos. O primeiro refere-se à decisão de escolha da marca, em que foi empregado o modelo logit condicional baseado na maximização da utilidade do domicílio. O segundo envolveu equações de demanda, obtidas pelo modelo clássico de regressão linear. Ambos foram especificados de forma que se pudesse testar a dependência das duas decisões de compra. No que diz respeito à validação, o modelo de escolha da marca demonstrou uma satisfatória capacidade de previsão, comparativamente aos modelos analisados na literatura. Implicações gerenciais incluem específicas decisões e ações de preço para as marcas, já que a natureza da decomposição das elasticidades-preço varia entre marcas.
Resumo:
O objetivo deste artigo é analisar o recente processo de fusões e a aquisições na economia brasileira, no tocante aos seus impactos sobre a concentração. Busca-se verificar correlações entre a natureza da operação e algumas variáveis econômicas referentes às empresas envolvidas. Através de dados dos relatórios de julgamento dos atos de concentração do Conselho Administrativo de Defesa Econômica (CADE) e dos pareceres econômicos formulados pela Secretaria de Acompanhamento Econômico do Ministério da Fazenda (SEAE), realiza-se um estudo econométrico através do modelo logit. A suposição é de que operações de natureza horizontal possuem efeitos prejudiciais maiores sobre a concentração de mercado do que as de natureza vertical e/ou conglomerado. O resultado obtido mostra que a desnacionalização de empresas e as operações ocorridas em alguns setores aumentam a concentração, enquanto que nas operações de abrangência mundial este efeito é menor.
Resumo:
Esse trabalho tem como objetivo estudar o comportamento da taxa de homicídio na população masculina e sua relação com variáveis econômicas nos estados de Minas Gerais, Rio de Janeiro e São Paulo entre 1981 e 1997. Nossa abordagem se diferencia do tratamento usual da literatura pela construção de taxas de homicídio específicas para cada idade entre 15 e 40 anos. As variáveis econômicas apresentam coeficientes significativamente diferente de zero para a população entre 15 e 19 anos. Como esperado, um aumento do salário real e uma queda da desigualdade reduzem a taxa de homicídio. Surpreendentemente, uma queda do desemprego parece aumentar a taxa de homicídio. A maior parte dos coeficientes, porém, converge para zero com o aumento da idade, tornando-se não significativos a partir dos 20 anos. Além disso, identificamos a existência de inércia nas taxas de homicídio: gerações com maior taxa de homicídio quando jovem tendem a apresentar maiores taxas de homicídio durante todo o restante do seu ciclo de vida. Dessa forma, se as variáveis econômicas induzem uma alta taxa de homicídio entre os jovens em determinado ano, essa taxa tende a permanecer elevada para a geração durante seu ciclo de vida independente do comportamento posterior da economia. Utilizamos, nessa análise, uma reformulação do tradicional modelo Logit que incorpora a variável dependente defasada.
Resumo:
This thesis aims to evaluate whether humorous television commercials (TVCs) work for non-prescription drugs, known as “over-the-counter” (OTC). The construct humor in advertising is controversial since it involves complex and broad typology, and depends on the audience characteristics. Several studies within different product categories indicated that some consumer goods are better suited for humorous TVCs, while others, such as OTC drugs, may not take advantage from it. Paradoxically, drug announcers spend billions of dollars worldwide in humorous OTC ads. An experiment with real consumers was designed as between-and-within-subjects, to test three hypotheses. Sixty women were exposed to pairs of humorous and non-humorous TVCs, for each of the three drug categories (analgesics, vitamins, and laxatives). We used fictional brand names and real ads, and measured four dependant variables: attitude toward the advertising (AAD), attitude toward the brand (ABR), purchase intention (PI), and brand choice (BC), after subjects being exposed to manipulations of two independent variables: humorous vs. non-humorous TV commercials, for the drug categories. Conditional logit model confirmed that humor does not help to persuade respondents, whose choices, attitudes, and purchase intention were less favorable with humorous TVCs, in comparison to non-humorous executions. Future research is presented regarding marketing for pharmaceutical products.
Resumo:
This paper studies the male homicide rate and its relation to economic variables in the states of Minas Gerais, São Paulo e Rio de Janeiro between 1981 and 1997. The novelty of our approach is the construction of homicide rates specific for each age between 15 and 40 years old. The economic variables' coefficients are significant1y different from zero for the population between 15 and 19 years old. As expected, an increase in real wage and a decrease in inequality reduce the rate of homicide. Surprisingly, a decrease in the unemployment rate seems to increase the rate ofhomicide. Most coefficients, however, converge to zero as a generation gets older, becoming non-significant for the population aged 20 years old or more. We also identify an inertia component in the homicide rate: generations with higher homicide rates when young also tend to have higher homicide rates over the remain of their life cycle. Therefore, if economic variables induce a high rate of homicide among young people in a certain year, this high rate tend to persist over the generation life cycle independent1y of the economy later behavior. Regressions are performed using a reformulation of the standard Logit model that incorporates a lagged dependent variable.
Resumo:
Labor force participation among youth is extremely high in Brazil when compared to countries with a similar economic background. In Argentina and Chile labor force participation, among those with 10 to 14 years old, is around 1% while in Brazil this rate is as high as 17 %. For the those between 15 and 19 years old these figures are around 10% in Chile, 15% in Argentina and 53% in Brazil. On the other hand the data on school attendance give a more optimistic picture. The percentage of children, between 10 and 14 years old, enrolled in school increased steadily from 79% to 95% from 1981 to 1998 and with age between 15 and 19, from 46% to 66% in the same period. These figures are close to the ones presented by Chile and Argentina. around 99% among the youngest group and around 70% for the 15 to 19 years old group. The objective of the paper is to understand the determinants of the time allocation decision of the Brazilian youth during the last twenty years. Using a multinomial logit regression we investigate the conditional effect of various micro and macro variables on the time allocation decision for the 1991 to 1998 period. Our main findings are: working and studying became the most likely allocation among the youngest in the poor rural areas and, in general, to study, whether working or not, became less dependent on family background for the youngest group but not for the older.
Resumo:
Neste trabalho buscamos identificar os principais determinantes da demanda por fundos de investimento no Brasil através do modelo Logit, que é bastante utilizado na teoria das organizações industriais. Sempre que possível realizamos “links” com os principais conceitos de finanças comportamentais. Assim, conseguimos aclarar as principais variáveis que impactam as variações de “market-share” na indústria de fundos de investimento. Concluímos que os principais indicadores observados pelos investidores no momento de tomada de decisão são o CDI, a inflação, a taxa real de juros, a variação do dólar e da bolsa de valores, por outro lado a rentabilidade acumulada dos últimos três meses é fator decisivo para que o investidor aplique ou resgate um fundo de investimento. Variáveis de risco e de retorno esperado que imaginávamos ter forte impacto, não se mostraram significativas para as variações de “share”.
Resumo:
Family farming has been considered as the new axis of rural development in the country, the focus of several public policies, especially the National Program for Strengthening Family Agriculture - PRONAF and Food Purchase Program - PAA. PRONAF was created with the aim of providing credit to farmers, while the PAA to support family farmers through the purchase of its production. In this context, the objective of this study is to analyze the correspondence of these two public policies for family farming, in the Territories of Citizenship of the state of Rio Grande do Norte, between the years 2008 to 2010. In the methodology, the analysis was performed by comparing the distributions of the two programs in the territories of citizenship status. There were also statistical tests of differences in proportions, and Spearman correlations, and estimated a logit regression model, in order to measure the probability of a farmer participating in the PAA is associated with one of the modes of PRONAF. The data used were obtained from the National and Supply - CONAB at the Institute of Technical Assistance and Rural Extension - EMATER, and the Ministry of Agrarian Development - MDA. Among the key findings was noted that policies were associated with a direct, but low in the districts of the Territories of Citizenship. And that, in the years 2008 and 2009, only in the territories of Mato Grande, Alto Oeste and Seridó, the actions of PAA and PRONAF had direct and significant correlations. It was found that in most of the territories, policies are performed randomly, ie that both have no correlation to each other. The estimates of the logit model showed that the chance of a family farmer, the PAA participant, receive credits PRONAF A, is higher in the territory of Mato Grande, and would have a chance to fall in PRONAF B in all areas surveyed. Moreover, farmers in the territories of the Assu-Mossoró, Sertão of Apodi, Seridó and Alto Oeste, participating in the PAA would be more likely to receive credits PRONAF C, reflecting thus the family farm more consolidated these territories