Optimal two-sided tests for instrumental variables regression with heteroskedastic and autocorrelated errors

This paper considers two-sided tests for the parameter of an endogenous variable in an instrumental variable (IV) model with heteroskedastic and autocorrelated errors. We develop the nite-sample theory of weighted-average power (WAP) tests with normal errors and a known long-run variance. We introduce two weights which are invariant to orthogonal transformations of the instruments; e.g., changing the order in which the instruments appear. While tests using the MM1 weight can be severely biased, optimal tests based on the MM2 weight are naturally two-sided when errors are homoskedastic. We propose two boundary conditions that yield two-sided tests whether errors are homoskedastic or not. The locally unbiased (LU) condition is related to the power around the null hypothesis and is a weaker requirement than unbiasedness. The strongly unbiased (SU) condition is more restrictive than LU, but the associated WAP tests are easier to implement. Several tests are SU in nite samples or asymptotically, including tests robust to weak IV (such as the Anderson-Rubin, score, conditional quasi-likelihood ratio, and I. Andrews' (2015) PI-CLC tests) and two-sided tests which are optimal when the sample size is large and instruments are strong. We refer to the WAP-SU tests based on our weights as MM1-SU and MM2-SU tests. Dropping the restrictive assumptions of normality and known variance, the theory is shown to remain valid at the cost of asymptotic approximations. The MM2-SU test is optimal under the strong IV asymptotics, and outperforms other existing tests under the weak IV asymptotics.

A study on time-varying quantile and its applications

This Thesis is the result of my Master Degree studies at the Graduate School of Economics, Getúlio Vargas Foundation, from January 2004 to August 2006. am indebted to my Thesis Advisor, Professor Luiz Renato Lima, who introduced me to the Econometrics' world. In this Thesis, we study time-varying quantile process and we develop two applications, which are presented here as Part and Part II. Each of these parts was transformed in paper. Both papers were submitted. Part shows that asymmetric persistence induces ARCH effects, but the LMARCH test has power against it. On the other hand, the test for asymmetric dynamics proposed by Koenker and Xiao (2004) has correct size under the presence of ARCH errors. These results suggest that the LM-ARCH and the Koenker-Xiao tests may be used in applied research as complementary tools. In the Part II, we compare four different Value-at-Risk (VaR) methodologies through Monte Cario experiments. Our results indicate that the method based on quantile regression with ARCH effect dominates other methods that require distributional assumption. In particular, we show that the non-robust method ologies have higher probability to predict VaRs with too many violations. We illustrate our findings with an empirical exercise in which we estimate VaR for returns of São Paulo stock exchange index, IBOVESPA, during periods of market turmoil. Our results indicate that the robust method based on quantile regression presents the least number of violations.

Purchasing power parity and the unit root tests: a robust analysis

Empirical evidence suggests that real exchange rate is characterized by the presence of near-unity and additive outliers. Recent studeis have found evidence on favor PPP reversion by using the quasi-differencing (Elliott et al., 1996) unit root tests (ERS), which is more efficient against local alternatives but is still based on least squares estimation. Unit root tests basead on least saquares method usually tend to bias inference towards stationarity when additive out liers are present. In this paper, we incorporate quasi-differencing into M-estimation to construct a unit root test that is robust not only against near-unity root but also against nonGaussian behavior provoked by assitive outliers. We re-visit the PPP hypothesis and found less evidemce in favor PPP reversion when non-Gaussian behavior in real exchange rates is taken into account.

Comparing value-at-risk methodologies

In this paper, we compare four different Value-at-Risk (V aR) methodologies through Monte Carlo experiments. Our results indicate that the method based on quantile regression with ARCH effect dominates other methods that require distributional assumption. In particular, we show that the non-robust methodologies have higher probability to predict V aRs with too many violations. We illustrate our findings with an empirical exercise in which we estimate V aR for returns of S˜ao Paulo stock exchange index, IBOVESPA, during periods of market turmoil. Our results indicate that the robust method based on quantile regression presents the least number of violations.

Modelagem do potencial de mercado dos municípios paulistas

Os indicadores de potencial de consumo são medidas indiretas da capacidade de uma região absorver uma determinada categoria de produto. O método de regressão linear simples pode ser utilizado na elaboração de índices simples e robustos que geralmente produzem resultados comparáveis ou superiores aos que podem ser obtidos pela utilização de índices vendidos por empresas de consultoria. Este fato passa despercebido pela maioria dos usuários porque habitualmente não são feitos esforços de comparar os índices com a realidade que estes se propõem a descrever (o que tem uma certa lógica, porque se conhecêssemos a realidade não precisaríamos do índice de potencial). Para estabelecer a metodologia e demonstrar a tese acima, são coletados dados a respeito da área de loja de supermercados dos municípios do Estado de São Paulo; um índice é construído, e em seguida comparado com outros indicadores. Os resultados confirmam a suposição inicial.

Electoral rules, political competition and fiscal spending: regression discontinuity evidence from brazilian municipalities

We exploit a discontinuity in Brazilian municipal election rules to investigate whether political competition has a causal impact on policy choices. In municipalities with less than 200,000 voters mayors are elected with a plurality of the vote. In municipalities with more than 200,000 voters a run-off election takes place among the top two candidates if neither achieves a majority of the votes. At a first stage, we show that the possibility of runoff increases political competition. At a second stage, we use the discontinuity as a source of exogenous variation to infer causality from political competition to fiscal policy. Our second stage results suggest that political competition induces more investment and less current spending, particularly personnel expenses. Furthermore, the impact of political competition is larger when incumbents can run for reelection, suggesting incentives matter insofar as incumbents can themselves remain in office.

On insurance markets with endogenous information acquisition: A robust mechanism design approach

Neste trabalho propomos a aplicação das noções de equilíbrio da recente literatura de desenho de mecanismo robusto com aquisição de informação endógena a um problema de divisão de risco entre dois agentes. Através deste exemplo somos capazes de motivar o uso desta noção de equilíbrio, assim como discutir os efeitos da introdu ção de uma restrição de participação que seja dependente da informação. A simplicidade do modelo nos permite caracterizar a possibilidade de implementar a alocação Pareto efiente em termos do custo de aquisição da informação. Além disso, mostramos que a precisão da informação pode ter um efeito negativo sobre a implementação da alocação efi ciente. Ao final, sao dados dois exemplos específicos de situações nas quais este modelo se aplica.

A robust poverty profile for Brazil using multiple data sources

This paper presents a poverty profile for Brazil, based on three different sources of household data for 1996. We use PPV consumption data to estimate poverty and indigence lines. “Contagem” data is used to allow for an unprecedented refinement of the country’s poverty map. Poverty measures and shares are also presented for a wide range of population subgroups, based on the PNAD 1996, with new adjustments for imputed rents and spatial differences in cost of living. Robustness of the profile is verified with respect to different poverty lines, spatial price deflators, and equivalence scales. Overall poverty incidence ranges from 23% with respect to an indigence line to 45% with respect to a more generous poverty line. More importantly, however, poverty is found to vary significantly across regions and city sizes, with rural areas, small and medium towns and the metropolitan peripheries of the North and Northeast regions being poorest.

O crime de drogas e a violência em São Paulo: uma análise a partir da Lei de Drogas

O objetivo desta pesquisa é estudar o efeito da Lei nº 11.343/06 (Lei de Drogas) sobre o crime de tráfico e porte de drogas e a relação entre crimes de drogas e outros crimes. Para isso, são exploradas as variações da Lei de Drogas, através de análises de regressões com descontinuidade e com variável instrumental, além de estimações com dados em painel, em busca de um efeito causal entre drogas e violência. Como resultados, a Lei de Drogas parece não ter efeito negativo significativo sobre crimes de drogas. Por outro lado, crimes de drogas apresentam uma associação negativa sobre crimes de furto e uma relação positiva com crimes de formação de quadrilha. Para cada redução de 100 crimes de drogas (por mil habitantes) associa-se um aumento de 3,6 crimes de furto (por mil habitantes) e uma diminuição de 27 crimes de formação de quadrilha (por mil habitantes). Não são encontrados efeitos robustos sobre roubos, homicídios, latrocínios, estupros, crimes de lesão corporal e porte de arma de fogo.

The Brazilian real state market in 2012: robust growth or speculative bubble?

A forte alta dos imóveis no Brasil nos últimos anos iniciou um debate sobre a possível existência de uma bolha especulativa. Dada a recente crise do crédito nos Estados Unidos, é factível questionar se a situação atual no Brasil pode ser comparada à crise americana. Considerando argumentos quantitativos e fundamentais, examina-se o contexto imobiliário brasileiro e questiona-se a sustentabilidade em um futuro próximo. Primeiramente, analisou-se a taxa de aluguel e o nível de acesso aos imóveis e também utilizou-se um modelo do custo real para ver se o mercado está em equilíbrio o não. Depois examinou-se alguns fatores fundamentais que afetam o preço dos imóveis – oferta e demanda, crédito e regulação, fatores culturais – para encontrar evidências que justificam o aumento dos preços dos imóveis. A partir dessas observações tentou-se chegar a uma conclusão sobre a evolução dos preços no mercado imobiliário brasileiro. Enquanto os dados sugerem que os preços dos imóveis estão supervalorizados em comparação ao preço dos aluguéis, há evidências de uma legítima demanda por novos imóveis na emergente classe média brasileira. Um risco maior pode estar no mercado de crédito, altamente alavancado em relação ao consumidor brasileiro. No entanto, não se encontrou evidências que sugerem mais do que uma temporária estabilização ou correção no preço dos imóveis.

The impact of OFDI on economic growth countries. An econometric approach using panel data and time-series evidence

The thesis at hand adds to the existing literature by investigating the relationship between economic growth and outward foreign direct investments (OFDI) on a set of 16 emerging countries. Two different econometric techniques are employed: a panel data regression analysis and a time-series causality analysis. Results from the regression analysis indicate a positive and significant correlation between OFDI and economic growth. Additionally, the coefficient for the OFDI variable is robust in the sense specified by the Extreme Bound Analysis (EBA). On the other hand, the findings of the causality analysis are particularly heterogeneous. The vector autoregression (VAR) and the vector error correction model (VECM) approaches identify unidirectional Granger causality running either from OFDI to GDP or from GDP to OFDI in six countries. In four economies causality among the two variables is bidirectional, whereas in five countries no causality relationship between OFDI and GDP seems to be present.

Wald tests for IV regression with weak instruments

This dissertation deals with the problem of making inference when there is weak identification in models of instrumental variables regression. More specifically we are interested in one-sided hypothesis testing for the coefficient of the endogenous variable when the instruments are weak. The focus is on the conditional tests based on likelihood ratio, score and Wald statistics. Theoretical and numerical work shows that the conditional t-test based on the two-stage least square (2SLS) estimator performs well even when instruments are weakly correlated with the endogenous variable. The conditional approach correct uniformly its size and when the population F-statistic is as small as two, its power is near the power envelopes for similar and non-similar tests. This finding is surprising considering the bad performance of the two-sided conditional t-tests found in Andrews, Moreira and Stock (2007). Given this counter intuitive result, we propose novel two-sided t-tests which are approximately unbiased and can perform as well as the conditional likelihood ratio (CLR) test of Moreira (2003).

Semiparametric estimation and inference using doubly robust moment conditions

We study semiparametric two-step estimators which have the same structure as parametric doubly robust estimators in their second step. The key difference is that we do not impose any parametric restriction on the nuisance functions that are estimated in a first stage, but retain a fully nonparametric model instead. We call these estimators semiparametric doubly robust estimators (SDREs), and show that they possess superior theoretical and practical properties compared to generic semiparametric two-step estimators. In particular, our estimators have substantially smaller first-order bias, allow for a wider range of nonparametric first-stage estimates, rate-optimal choices of smoothing parameters and data-driven estimates thereof, and their stochastic behavior can be well-approximated by classical first-order asymptotics. SDREs exist for a wide range of parameters of interest, particularly in semiparametric missing data and causal inference models. We illustrate our method with a simulation exercise.

Uncertainty of aid inflows and the growth relationship

This paper contributes to the literature on aid and economic growth. We posit that it is not the levei of aid flows per se but the stability of such flows that determines the impact of aid on economic growth. Three measures of aid instability are employed. One is a simple deviation from trend, and measures overall instability. The other measures are based on auto-regressive estimates to capture deviations from an expected trend. These measures are intended to proxy for uncertainty in aid receipts. We posit that such uncertainty will influence the relationship between aid and investment and how recipient governments respond to aid, and will therefore affect how aid impacts on growth. We estimate a standard cross-country growth regression including the leveI of aid, and find aid to be insignificant (in line with other results in the literature). We then introduce measures of instability. Aid remains insignificant when we account for overall instability. However, when we account for uncertainty (which is negative and significant), we find that aid has a significant positive effect on growth. We conduct stability tests that show that the significance of aid is largely due to its effect on the volume of investment. The finding that uncertainty of aid receipts reduces the effectiveness of aid is robust. When we control for this, aid appears to have a significant positive influence on growth. When the regression is estimated for the sub-sample of African countries these findings hold, although the effectiveness of aid appears weaker than for the full sample.