Probability matching is the tendency to predict outcomes in accordance with their actual contingencies in a binary choice task. It is, however, a suboptimal response if the aim is to maximize correct predictions. I review two theories that attempt to explain why probability matching occurs: the pattern-search hypothesis and dual-systems theory. These theories are tested in two studies which suggest that dual-systems theory provides a better account of probability matching behavior. Studies 3, 4, and 5 then provide evidence for an extension of the dual-systems theory, called expectation matching, which is intended to explain why probability matching is the intuitive response to a binary choice problem.

Probability matching is the tendency to predict outcomes in accordance with their actual contingencies in a binary choice task. It is, however, a suboptimal response if the aim is to maximize correct predictions. I review two theories that attempt to explain why probability matching occurs: the pattern-search hypothesis and dual-systems theory. These theories are tested in two studies which suggest that dual-systems theory provides a better account of probability matching behavior. Studies 3, 4, and 5 then provide evidence for an extension of the dual-systems theory, called expectation matching, which is intended to explain why probability matching is the intuitive response to a binary choice problem.

This thesis develops a conceptual framework to determine the probability of adopting HT canola when producers are assumed heterogeneous. The model is based on the framework developed by Fulton and Keyowski (1999), but is modified from a deterministic model to a probabilistic model. The study also considers the gross returns from adopting HT canola. Canola production in Manitoba, Canada is chosen as the region of analysis for the empirical component of the study. In 2002, 74 per cent of total canola acres in Manitoba were devoted to HT canola production. Factors such as soil type, producer risk profile, experience, productivity, and management ability are considered as potential determining factors which distinguish adopters of HT technology from non-adopters. Based on an initial assessment of Manitoba canola data, which shows the incomplete adoption of HT technology in Manitoba, a model is developed which considers adoption of a new technology as a function of the characteristics of the adopters. The conceptual model is tested empirically in two-stages. The first stage employs Ordinary Least Squares analysis to estimate the expected yield of different canola varieties to determine whether producers realize a benefit from the adoption of HT varieties. A logit analysis is conducted in the second stage, and considers different attributes of producers such as risk aversion, management ability, productivity and expected yields to determine the probability of producers adopting HT technology. The results show two primary findings. First, certain HT varieties can be shown to give producers higher returns. Second, differentiating characteristics of producers are key in determining the likely adoption of HT canola.

This thesis develops a conceptual framework to determine the probability of adopting HT canola when producers are assumed heterogeneous. The model is based on the framework developed by Fulton and Keyowski (1999), but is modified from a deterministic model to a probabilistic model. The study also considers the gross returns from adopting HT canola. Canola production in Manitoba, Canada is chosen as the region of analysis for the empirical component of the study. In 2002, 74 per cent of total canola acres in Manitoba were devoted to HT canola production. Factors such as soil type, producer risk profile, experience, productivity, and management ability are considered as potential determining factors which distinguish adopters of HT technology from non-adopters. Based on an initial assessment of Manitoba canola data, which shows the incomplete adoption of HT technology in Manitoba, a model is developed which considers adoption of a new technology as a function of the characteristics of the adopters. The conceptual model is tested empirically in two-stages. The first stage employs Ordinary Least Squares analysis to estimate the expected yield of different canola varieties to determine whether producers realize a benefit from the adoption of HT varieties. A logit analysis is conducted in the second stage, and considers different attributes of producers such as risk aversion, management ability, productivity and expected yields to determine the probability of producers adopting HT technology. The results show two primary findings. First, certain HT varieties can be shown to give producers higher returns. Second, differentiating characteristics of producers are key in determining the likely adoption of HT canola.

Thesis advisor: Arthur Lewbel / My doctoral research comprises both structural empirical work in the field of household economics (Chapters 1 and 3) and reduced-form theoretical work in econometrics (Chapter 2). The first chapter quantifies consumption gains in the form of consumption expenditure savings for couples living together versus apart. Moreover, I comprehensively analyze the key drivers of heterogeneity in consumption gains across couples. The second chapter, which is joint work with Stefan Hoderlein, develops the first difference-in-differences model for binary choice outcome variables when treatment effects are heterogeneous. Finally, in the third chapter, I study the realized gains in consumption levels upon move-in from single to couple and move-out from couple to single, separately by gender. Chapter 1. In “Heterogeneity in Consumption Gains from Living Together as a Couple”, I estimate consumption gains from living together versus apart. Specifically, this chapter quantifies these gains from consumption economies of scale for married as well as unmarried cohabiting couples in the U.S. I also comprehensively analyze the determinants of heterogeneity in consumption gains. For this purpose, I extend a recent collective household model to incorporate heterogeneity in consumption economies of scale and perform structural estimation using data from the Panel Study of Income Dynamics (PSID). I find that couples living together save an average of 14 percent in annual consumption expenditures compared to their counterfactual spending when living apart. Moreover, older couples and homeowner couples experience higher consumption economies of scale and thus higher consumption gains. I do not detect marital status to be a determining factor. These results have implications for appropriate adjustment of poverty thresholds across households of different sizes. Chapter 2. In “Binary Choice Difference-in-Differences Model with Heterogeneous Treatment Effects”, we develop the first Difference-in-Differences Model for binary choice outcome variables when treatment effects are heterogeneous. Our main result establishes identification of the average treatment effect on the treated (ATT). Moreover, we present two extensions. First, we show identification of the joint distribution of the actual and counterfactual latent outcome variable in the treatment group. Second, we extend the basic model to allow for the inclusion of covariates. We suggest an estimator for the ATT and evaluate its finite sample properties with the help of Monte Carlo simulations. Chapter 3. In “Realized Consumption Gains from Living Together as a Couple by Gender”, I am among the first to analyze how consumption gains from living together as a couple are split between men and women and whether this varies by individual-level or couple-level observables. Understanding individual-level consumption gains is important for rationalizing marriage and cohabitation decisions on the micro-level, and marriage and cohabitation patterns on the macro-level. I use the extended collective household model from Chapter 1 and data from the PSID to estimate male and female realized consumption gains from moving in as a couple and moving out of a couple. Average realized consumption gains upon move-in are positive for women and negative for men. This reverses for dissolving couples. Lastly, individuals who move in with their spouse have higher realized consumption gains than individuals who decide to cohabit without a marriage certificate. / Thesis (PhD) — Boston College, 2018. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Economics.

Estudos sobre a relação da poluição atmosférica e saúde apontam que há relação entre as queimadas de cana-de-açúcar e internações hospitalares, gerando custos para a sociedade. Este trabalho apresenta uma aplicação do Método de Avaliação Contingente para valorar os custos do bem-estar causados pela queimada de cana-de-açúcar. O fogo utilizado em um largo processo gera vários problemas para os moradores, passando por problemas de saúde pelo inconveniente causado pela fuligem que cai sobre as cidades no período de abril a novembro. A aplicação é feita na cidade de Ribeirão Preto, um dos maiores produtores de cana-de-açúcar e álcool no Brasil. Por meio dos métodos logit censurado, Heckman em dois estágios adaptado e Máxima Verossimilhança, calculou-se a disposição a pagar média e os resultados obtidos indicam que o custo estimado é em torno de R$ 180 milhões, para o período de 2009 a 2017. Tal valor está provavelmente subestimado já que o custo do bem-estar envolvendo as queimadas de cana-de-açúcar também passa pelos gastos com as internações hospitalares por doenças respiratórias. Dessa forma, propôs-se a verificação da relação entre as internações e a presença de cana-de-açúcar nos municípios do Estado de São Paulo. De acordo com dados do DATASUS, a região Sudeste gasta em torno de R$ 20 milhões por ano com esse tipo de internação, porém esse custo também está subestimado já que não leva em consideração os custos com inalação, atendimento ambulatorial e outros tratamentos posteriores à internação. / Studies on the relation of air pollution and health said that there is a relationship between sugar cane burning and hospital attendance, generating costs to society. This work presents an application of Contingent Valuation Method to evaluate the welfare costs of sugar cane fire. The use of fire in the process of sugar cane leads to many problems to near fire residents such as health problems caused by the soot over the cities from April to November. The application was made for the city of Ribeirão Preto, the major city in production of sugar cane and alcohol in Brazil. Using Censored Logit, Two Stage Heckman adapted and Maximum Likelihood methods an average willing to pay was estimated and the results indicate that the estimated cost is about R$ 180 millions from 2009 to 2017. This estimated value probably is underestimated since welfare costs of sugar cane burning for health also involve spending in hospitals admissions and respiratory diseases. Therefore, it is verified the relation of hospital admissions and the presence of sugar cane in São Paulo State cities. Using DATASUS database the Southeast region of Brazil spends R$ 20 millions a year with this type of hospital admissions, however this cost is also underestimated because it doesn\'t account for inhalation costs, ambulatory services and posterior treatment due admission.

nÃo hà / The paper develops a tool for modeling the bank credit risk and applies this to banking market of Fortaleza. Using data from a large commercial bank of the city for 290 customers with active accounts and minimum income of six hundred reais, were selected 23 control variables and was estimated the probability of default on the modalities check and other credit restrictions. The results showed that: i) females are less likely to face restrictions, although this is not a determinant of emissions of bad checks; ii) people who have insurance contracted with the bank showed themselves more likely to default and iii) the extent of the bank rating proposal was effective in measuring the chance of credit risk. / O trabalho desenvolve uma ferramenta para modelar o risco de crÃdito bancÃrio e aplica ao mercado bancÃrio de Fortaleza. A partir de dados de um grande banco comercial da cidade para 290 clientes com contas ativas e renda mÃnima de seiscentos reais, foram selecionadas 23 variÃveis de controle e estimou-se a probabilidade de inadimplÃncia nas modalidades cheque e demais restriÃÃes de crÃdito. Os resultados demonstram que: i) indivÃduos do sexo feminino possuem menos chance de enfrentar restriÃÃes, muito embora este nÃo seja um determinante das emissÃes de cheques sem fundos; ii) os indivÃduos que possuem seguro contratado junto ao banco apresentaram maior chance de inadimplÃncia e iii) a medida de rating proposta pelo banco se mostrou eficaz em mensurar a chance de risco de crÃdito.

Estudos sobre a relação da poluição atmosférica e saúde apontam que há relação entre as queimadas de cana-de-açúcar e internações hospitalares, gerando custos para a sociedade. Este trabalho apresenta uma aplicação do Método de Avaliação Contingente para valorar os custos do bem-estar causados pela queimada de cana-de-açúcar. O fogo utilizado em um largo processo gera vários problemas para os moradores, passando por problemas de saúde pelo inconveniente causado pela fuligem que cai sobre as cidades no período de abril a novembro. A aplicação é feita na cidade de Ribeirão Preto, um dos maiores produtores de cana-de-açúcar e álcool no Brasil. Por meio dos métodos logit censurado, Heckman em dois estágios adaptado e Máxima Verossimilhança, calculou-se a disposição a pagar média e os resultados obtidos indicam que o custo estimado é em torno de R$ 180 milhões, para o período de 2009 a 2017. Tal valor está provavelmente subestimado já que o custo do bem-estar envolvendo as queimadas de cana-de-açúcar também passa pelos gastos com as internações hospitalares por doenças respiratórias. Dessa forma, propôs-se a verificação da relação entre as internações e a presença de cana-de-açúcar nos municípios do Estado de São Paulo. De acordo com dados do DATASUS, a região Sudeste gasta em torno de R$ 20 milhões por ano com esse tipo de internação, porém esse custo também está subestimado já que não leva em consideração os custos com inalação, atendimento ambulatorial e outros tratamentos posteriores à internação. / Studies on the relation of air pollution and health said that there is a relationship between sugar cane burning and hospital attendance, generating costs to society. This work presents an application of Contingent Valuation Method to evaluate the welfare costs of sugar cane fire. The use of fire in the process of sugar cane leads to many problems to near fire residents such as health problems caused by the soot over the cities from April to November. The application was made for the city of Ribeirão Preto, the major city in production of sugar cane and alcohol in Brazil. Using Censored Logit, Two Stage Heckman adapted and Maximum Likelihood methods an average willing to pay was estimated and the results indicate that the estimated cost is about R$ 180 millions from 2009 to 2017. This estimated value probably is underestimated since welfare costs of sugar cane burning for health also involve spending in hospitals admissions and respiratory diseases. Therefore, it is verified the relation of hospital admissions and the presence of sugar cane in São Paulo State cities. Using DATASUS database the Southeast region of Brazil spends R$ 20 millions a year with this type of hospital admissions, however this cost is also underestimated because it doesn\'t account for inhalation costs, ambulatory services and posterior treatment due admission.

Energy production is associated with environmental impairment. Most anxious is the greenhouse gas emissions, which also arise from transportation. If battery electric vehicles should be able to alleviate the problem, they must be charged with environmentally friendly produced electricity. This paper investigates a possible relationship between battery electric vehicles and solar photovoltaic panels in household survey data from ENABLE.EU performed in ten European countries autumn 2017 – spring 2018. Estimated with a logit binary choice model, it is found that the probability that a household owns a battery electric vehicle increases if the household owns solar photovoltaic panels. Furthermore, this increase in probability is higher within countries with a higher market diffusion of battery electric vehicle and solar photovoltaic panels (France and the UK). This suggests that policy encouraging home charging of battery electric vehicles from solar photovoltaic panels that includes an energy storage facility could speed up the transition of the vehicle fleet.

Made available in DSpace on 2016-04-26T20:48:41Z (GMT). No. of bitstreams: 1 Danilo Marques de Mendonca.pdf: 845535 bytes, checksum: 668ae8796a8ad81784166717cbbe69bc (MD5) Previous issue date: 2014-05-28 / After the period of monetary stabilization started with the Real Plan in 1994 , the credit market has shown annual growth rates of 20 %. About 40 % of this growth came from the credit market for individuals . This paper analyzed the profile of the families who have credit expenses, and what changes in their characteristics can cause any effect in their propensity to take credit . For this purpose we applied binary logit choice model based on microdata from the Household Budget Survey (POF 2008 / 09 ) of the IBGE, in an attempt to measure the probability of the family take a loan. For this, we used categorical variables relating to the constitution of families, such as education level, sex, race and age of household head, and other information on the composition of household expenditures found in POF. The data suggest that the two most important factors to increase the likelihood of family borrowing is the age of the household head and income per capita. However other factors also contribute significantly, such as the existence of financial investment spending , expend with reform the household or even health spending, children's age, sex, race and education of household head / Após o período de estabilização monetária iniciado com o Plano Real em 1994, o mercado de crédito brasileiro vem apresentando taxas de crescimento anuais nominais acima de 20%. Cerca de 40% deste crescimento advêm do mercado de crédito direcionado às pessoas físicas. Neste trabalho é analisado o perfil das famílias que possuem despesas com crédito, e quais mudanças em suas características podem causar alterações em sua propensão a tomar crédito. Para tal objetivo foi aplicado o modelo de escolha binária logit à base dos microdados da Pesquisa de Orçamento Familiar (POF 2008/09) do IBGE, na tentativa de mensurar a probabilidade da família ser tomadora de crédito. Para tanto, são usadas variáveis categóricas referentes à constituição das famílias, como: grau de escolaridade, sexo, raça e idade do chefe da família, além de outras informações sobre a composição das despesas familiares encontradas na POF. Os dados sugerem que os fatores mais relevantes a aumentar a probabilidade da família tomar empréstimos são a idade do chefe da família e a renda per capita. No entanto outros fatores também contribuem significativamente, tais como a existência de gastos com aplicação financeira, gastos com reforma do domicílio ou mesmo com saúde emergencial, idade dos filhos, sexo, raça e educação do chefe da família