Essays on Spatial Econometrics

Esta disserta����o concentra-se nos processos estoc��sticos espaciais definidos em um reticulado, os chamados modelos do tipo Cliff & Ord. Minha contribui����o nesta tese consiste em utilizar aproxima����es de Edgeworth e saddlepoint para investigar as propriedades em amostras finitas do teste para detectar a presen��a de depend��ncia espacial em modelos SAR (autoregressivo espacial), e propor uma nova classe de modelos econométricos espaciais na qual os par��metros que afetam a estrutura da m��dia s��o distintos dos par��metros presentes na estrutura da vari��ncia do processo. Isto permite uma interpreta����o mais clara dos par��metros do modelo, al��m de generalizar uma proposta de taxonomia feita por Anselin (2003). Eu proponho um estimador para os par��metros do modelo e derivo a distribui����o assint��tica do estimador. O modelo sugerido na disserta����o fornece uma interpreta����o interessante ao modelo SARAR, bastante comum na literatura. A investiga����o das propriedades em amostras finitas dos testes expande com rela����o a literatura permitindo que a matriz de vizinhan��a do processo espacial seja uma fun����o n��o-linear do par��metro de depend��ncia espacial. A utiliza����o de aproxima����es ao inv��s de simula����es (mais comum na literatura), permite uma maneira f��cil de comparar as propriedades dos testes com diferentes matrizes de vizinhan��a e corrigir o tamanho ao comparar a pot��ncia dos testes. Eu obtenho teste invariante ��timo que �� tamb��m localmente uniformemente mais potente (LUMPI). Construo o envelope de pot��ncia para o teste LUMPI e mostro que ele �� virtualmente UMP, pois a pot��ncia do teste est�� muito pr��xima ao envelope (considerando as estruturas espaciais definidas na disserta����o). Eu sugiro um procedimento pr��tico para construir um teste que tem boa pot��ncia em uma gama de situa����es onde talvez o teste LUMPI n��o tenha boas propriedades. Eu concluo que a pot��ncia do teste aumenta com o tamanho da amostra e com o par��metro de depend��ncia espacial (o que est�� de acordo com a literatura). Entretanto, disputo a vis��o consensual que a pot��ncia do teste diminui a medida que a matriz de vizinhan��a fica mais densa. Isto reflete um erro de medida comum na literatura, pois a dist��ncia estat��stica entre a hip��tese nula e a alternativa varia muito com a estrutura da matriz. Fazendo a corre����o, concluo que a pot��ncia do teste aumenta com a dist��ncia da alternativa �� nula, como esperado.

Modelo de estresse macroecon��mico da inadimpl��ncia para bancos de atacado

O presente trabalho visa propor uma metodologia de c��lculo de estresse para cumprir as exig��ncias regulat��rias do Bank for International Settlements (BIS) e Comiss��o de Supervis��o Banc��ria. A metodologia abordada utiliza as vari��veis macroecon��micas para se determinar o comportamento da inadimpl��ncia de uma determinada carteira de cr��dito de uma institui����o financeira. Para isso, foi dividida em dois est��gios: o primeiro modelo responde como seria a inadimpl��ncia dada as vari��veis macroecon��micas; o segundo modelo equilibra e correlaciona, proporcionalmente, as vari��veis macroecon��micas entre si e suas respostas a um choque econ��mico. Com os dois modelos alinhados �� poss��vel simular cen��rios positivos e negativos e saber os pontos de m��ximo da inadimpl��ncia. Assim, as institui����es financeiras podem determinar melhor suas pol��ticas de gest��o de risco e retorno.

Credit borrowing constraints in a DSGE framework: income vs. housing

Credit market in Brazil distinguishes from advanced economies in many aspects. One of them is related to collaterals for households borrowing. This work proposes a DSGE framework, based on Gerali et al.(2010), to analyse one pecularity of Brazillian credit market: payroll-deducted personal loans. To original model, we added the possibility to households contract long term debt and compare to differents types of credit constrains: one based on housing and other based on future income. We callibrate and estimate the model to Brazil, using Bayesian technique. Results show that, in a economy where credit constraints are based on income, responses to shocks appear to be stronger, at first, but dissipate faster. This occurs because income responds quickly to shock than housing prices, so does amount available to loans. In order to smooth consumption, agents compensate lower income and borrowing by increasing working hours, restoring loans and debt in a shorter time.

Ensaios em econometria aplicada

Este tese �� composta por quatro ensaios sobre aplica����es econom��tricas em t��picos econ��micos relevantes. Os estudos versam sobre consumo de bens n��o-dur��veis e pre��os de im��veis, capital humano e crescimento econ��mico, demanda residencial de energia el��trica e, por fim, periodicidade de vari��veis fiscais de Estados e Munic��pios brasileiros. No primeiro artigo, "Non-Durable Consumption and Real-Estate Prices in Brazil: Panel-Data Analysis at the State Level", �� investigada a rela����o entre varia����o do pre��o de im��veis e varia����o no consumo de bens n��o-dur��veis. Os dados coletados permitem a forma����o de um painel com sete estados brasileiros observados entre 2008- 2012. Os resultados s��o obtidos a partir da estima����o de uma forma reduzida obtida em Campbell e Cocco (2007) que aproxima um modelo estrutural. As estimativas para o caso brasileiro s��o inferiores as de Campbell e Cocco (2007), que, por sua vez, utilizaram microdados brit��nicos. O segundo artigo, "Uma medida alternativa de capital humano para o estudo emp��rico do crescimento", prop��e uma forma de mensura����o do estoque de capital humano que reflita diretamente pre��os de mercado, atrav��s do valor presente do fluxo de renda real futura. Os impactos dessa medida alternativa s��o avaliados a partir da estima����o da fun����o de produ����o tradicional dos modelos de crescimento neocl��ssico. Os dados comp��em um painel de 25 pa��ses observados entre 1970 e 2010. Um exerc��cio de robustez �� realizado para avaliar a estabilidade dos coeficientes estimados diante de varia����es em vari��veis ex��genas do modelo. Por sua vez, o terceiro artigo "Household Electricity Demand in Brazil: a microdata approach", parte de dados da Pesquisa de Or��amento Familiar (POF) para mensurar a elasticidade pre��o da demanda residencial brasileira por energia el��trica. O uso de microdados permite adotar abordagens que levem em considera����o a sele����o amostral. Seu efeito sobre a demanda de eletricidade �� relevante, uma vez que esta demanda �� derivada da demanda por estoque de bens dur��veis. Nesse contexto, a escolha pr��via do estoque de bens dur��veis (e consequentemente, a escolha pela intensidade de energia desse estoque) condiciona a demanda por eletricidade dos domic��lios. Finalmente, o quarto trabalho, "Interpola����o de Vari��veis Fiscais Brasileiras usando Representa����o de Espa��o de Estados" procurou sanar o problema de baixa periodicidade da divulga����o de s��ries fiscais de Estados e Munic��pios brasileiros. Atrav��s de t��cnica de interpola����o baseada no Filtro de Kalman, as s��ries mensais n��o observadas s��o projetadas a partir de s��ries bimestrais parcialmente observadas e covari��veis mensais selecionadas.

Forecasting Multivariate Time Series under Present-Value-Model Short- and Long-run Co-movement Restrictions

This paper has two original contributions. First, we show that the present value model (PVM hereafter), which has a wide application in macroeconomics and fi nance, entails common cyclical feature restrictions in the dynamics of the vector error-correction representation (Vahid and Engle, 1993); something that has been already investigated in that VECM context by Johansen and Swensen (1999, 2011) but has not been discussed before with this new emphasis. We also provide the present value reduced rank constraints to be tested within the log-linear model. Our second contribution relates to forecasting time series that are subject to those long and short-run reduced rank restrictions. The reason why appropriate common cyclical feature restrictions might improve forecasting is because it finds natural exclusion restrictions preventing the estimation of useless parameters, which would otherwise contribute to the increase of forecast variance with no expected reduction in bias. We applied the techniques discussed in this paper to data known to be subject to present value restrictions, i.e. the online series maintained and up-dated by Shiller. We focus on three different data sets. The fi rst includes the levels of interest rates with long and short maturities, the second includes the level of real price and dividend for the S&P composite index, and the third includes the logarithmic transformation of prices and dividends. Our exhaustive investigation of several different multivariate models reveals that better forecasts can be achieved when restrictions are applied to them. Moreover, imposing short-run restrictions produce forecast winners 70% of the time for target variables of PVMs and 63.33% of the time when all variables in the system are considered.

Mem��ria longa em dados intradi��rios: um estudo sobre proje����es baseadas na ordem fracion��ria de integra����o dos Retornos de A����es e ��ndices de a����es

Mandelbrot (1971) demonstrou a import��ncia de considerar depend��ncias de longo prazo na precifica����o de ativos - o m��todo tradicional para mensur��-las, encontrado em Hurst (1951), faz uso da estat��stica R/S. Paralelamente a isso, Box e Jenkins (1976; edi����o original de 1970) apresentaram sua famosa metodologia para determina����o da ordem dos par��metros de modelos desenvolvidos no contexto de processos com mem��ria de curto prazo, conhecidos por ARIMA (acr��nimo do ingl��s Autoregressive Integrated Moving Average). Estimulados pela percep����o de que um modelo que pretenda representar fielmente o processo gerador de dados deva explicar tanto a din��mica de curto prazo quanto a de longo prazo, Granger e Joyeux (1980) e Hosking (1981) introduziram os modelos ARFIMA (de onde o F adicionado vem de Fractionally), uma generaliza����o da classe ARIMA, nos quais a depend��ncia de longo prazo estimada �� relacionada ao valor do par��metro de integra����o. Pode-se dizer que a partir de ent��o processos com alto grau de persist��ncia passaram a atrair cada vez mais o interesse de pesquisadores, o que resultou no desenvolvimento de outros m��todos para estim��-la, por��m sem que algum tenha se sobressa��do claramente ��� e �� neste ponto que o presente trabalho se insere. Por meio de simula����es, buscou-se: (1) classificar diversos estimadores quanto a sua precis��o, o que nos obrigou a; (2) determinar parametriza����es razo��veis desses, entendidas aqui como aquelas que minimizam o vi��s, o erro quadr��tico m��dio e o desvio-padr��o. Ap��s rever a literatura sobre o tema, abordar estes pontos se mostrou necess��rio para o objetivo principal: elaborar estrat��gias de negocia����o baseadas em proje����es feitas a partir da caracteriza����o de depend��ncias em dados intradi��rios, minuto a minuto, de a����es e ��ndices de a����es. Foram analisadas as s��ries de retornos da a����o Petrobras PN e do ��ndice Bovespa, com dados de 01/04/2013 a 31/03/2014. Os softwares usados foram o S-Plus e o R.

An empirical comparison of forward-rate and spot-rate models for valuing interest-rate options

Our main goal is to investigate the question of which interest-rate options valuation models are better suited to support the management of interest-rate risk. We use the German market to test seven spot-rate and forward-rate models with one and two factors for interest-rate warrants for the period from 1990 to 1993. We identify a one-factor forward-rate model and two spot-rate models with two faetors that are not significant1y outperformed by any of the other four models. Further rankings are possible if additional cri teria are applied.

The forward and the equity premium puzzles: two symptoms of the same illness?

The Forward Premium Puzzle (FPP) is how the empirical observation of a negative relation between future changes in the spot rates and the forward premium is known. Modeling this forward bias as a risk premium and under weak assumptions on the behavior of the pricing kernel, we characterize the potential bias that is present in the regressions where the FPP is observed and we identify the necessary and sufficient conditions that the pricing kernel has to satisfy to account for the predictability of exchange rate movements. Next, we estimate the pricing kernel applying two methods: i) one, du.e to Ara��jo et aI. (2005), that exploits the fact that the pricing kernel is a serial correlation common feature of asset prices, and ii) a traditional principal component analysis used as a procedure 1;0 generate a statistical factor modeI. Then, using on the sample and out of the sample exercises, we are able to show that the same kernel that explains the Equity Premi um Puzzle (EPP) accounts for the FPP in all our data sets. This suggests that the quest for an economic mo deI that generates a pricing kernel which solves the EPP may double its prize by simultaneously accounting for the FPP.

Pricing the option adjust spread of Brazilian Eurobonds

This paper presents results of a pricing system to compute the option adjusted spread ("DAS") of Eurobonds issued by Brazilian firms. The system computes the "DAS" over US treasury rates taktng imo account the embedded options present on these bonds. These options can be calls ("callable bond"), puts ("putable bond") or combinations ("callable and putable bond"). The pricing model takes into account the evolution of the term structure along time, is compatible with the observable market term structure and is able to compute risk measures such as duration and convexity, and pricing and hedging of options on these bonds. Examples show the ejJects of the embedded options on the spread and risk measures as well as the ejJects on the spread due to variations on the volatility parameters ofthe short rate.

Does monetary policy stabilize the exchange rate following a currency crisis?

This paper provides evidence on the relationship between rnonetary policy and the exchange rate in the aftermath of currency crises. It ana1yzes a large data set of currency crises in 80 countries in the period 1980 to 1998. The rnain question addressed is: can rnonetary policy significantly alter the probability of reversing the post-crisis undervaluation through nominal appreciation rather than higher int1ation? We find that tight rnonetary policy facilitates the reversal of currency undervaluation through nominal appreciation rather than inflation. When the econorny is also facing a banking crisis, depending on the specification, tight rnonetary policy rnay not have the same effect.

Interests prices and the Barosky - Summers resolution of the Gibson paradox under the gold standard

This paper presents a structural monetary ��amework featunng a demand function for non-monetary uses of gold, such as the one drawn by Barsky and Summers in their 1988 analy8��I of the Gibson Paradox as a natural concomitant of the gold standard period. That structural model predicts that the laws of behavior of nominal prices and interest rates are functions of the rules set by the government to command the money supply. !ta fiduciary vemon obtaina Fisherian relationships &8 particular cases. !ta gold atandard 801ution yields a modelsimilar to the Barsky and Summers model, in which interest rates are exogeneous and subject to shocb. This paper integrates governnment bonds into the analysis, treats interest rates endogenously, and ahifts the responsibility for the shocb to the government budgetary financing policies. The Gibson paradox appears as "practically" the only cl&18 of behavioral pattern open for interest rates and price movements under apure gold standard economy. Fisherian-like relationshipe are utterly ruled out.

Market clearing sealed-bid auctions for non-identical objects with single-unit demands

We consider a version of the cooperative buyer-seller market game of Shapley and Shubik (1972). For this market we propose a c1ass of sealed- bid auctions where objects are sold simultaneously at a market c1earing price rule. We ana1yze the strategic games induced by these mechanisms under the complete information approach. We show that these noncooperative games can be regarded as a competitive process for achieving a cooperative outcome: every Nash equilibrium payoff is a core outcome of the cooperative market game. Precise answers can be given to the strategic questions raised.

Trade liberalization and the evolution of skill earnings differentials in Brazil

From 1988 to 1995, when trade liberalization was implemented in Brazil, relative earnings of skilled workers decreased. In this paper, we investigate the role of trade liberalization in explaining these relative earnings movements, by checking all the steps predicted by the HeckscherOhlin- style trade transmission mechanism. We find that: i) employment shifted from skilled to unskilled intensive sectors, and each Sector increased its relative share of skilled labor; ii) relative prices fell in skill intensive sectors; iii) tariff changes across sectors were not related to skill intensities, but the pass-through from tariffs to prices was stronger in skill intensive sectors; iv) the decline in skilled eamings differentials mandated by the price variation predicted by trade is very elose to the observed one. The results are compatible with trade liberalization, accounting for the observed rei ative eamings changes in Brazil.

A pricing model for sovereign bonde

Similar to the modeling used to evaluate ccnporate boncls, where it is a put optioo. 011 corporate assets, we modeled sovereign bonds. Instead of company's assets as underlining assets, we used foreign excbange reserves. The results show a fundamental pricing model for sovereign bond and an optimum relation between the debt size, term, mix between floating and fixed interest payments, and size of reserves. The model is tested with a Brazilian BradyBond.