Finite-Sample Diagnostics for Multivariate Regressions with Applications to Linear Asset Pricing Models

In this paper, we propose several finite-sample specification tests for multivariate linear regressions (MLR) with applications to asset pricing models. We focus on departures from the assumption of i.i.d. errors assumption, at univariate and multivariate levels, with Gaussian and non-Gaussian (including Student t) errors. The univariate tests studied extend existing exact procedures by allowing for unspecified parameters in the error distributions (e.g., the degrees of freedom in the case of the Student t distribution). The multivariate tests are based on properly standardized multivariate residuals to ensure invariance to MLR coefficients and error covariances. We consider tests for serial correlation, tests for multivariate GARCH and sign-type tests against general dependencies and asymmetries. The procedures proposed provide exact versions of those applied in Shanken (1990) which consist in combining univariate specification tests. Specifically, we combine tests across equations using the MC test procedure to avoid Bonferroni-type bounds. Since non-Gaussian based tests are not pivotal, we apply the “maximized MC” (MMC) test method [Dufour (2002)], where the MC p-value for the tested hypothesis (which depends on nuisance parameters) is maximized (with respect to these nuisance parameters) to control the test’s significance level. The tests proposed are applied to an asset pricing model with observable risk-free rates, using monthly returns on New York Stock Exchange (NYSE) portfolios over five-year subperiods from 1926-1995. Our empirical results reveal the following. Whereas univariate exact tests indicate significant serial correlation, asymmetries and GARCH in some equations, such effects are much less prevalent once error cross-equation covariances are accounted for. In addition, significant departures from the i.i.d. hypothesis are less evident once we allow for non-Gaussian errors.

Time-series regression models to study the short-term effects of environmental factors on health

Time series regression models are especially suitable in epidemiology for evaluating short-term effects of time-varying exposures on health. The problem is that potential for confounding in time series regression is very high. Thus, it is important that trend and seasonality are properly accounted for. Our paper reviews the statistical models commonly used in time-series regression methods, specially allowing for serial correlation, make them potentially useful for selected epidemiological purposes. In particular, we discuss the use of time-series regression for counts using a wide range Generalised Linear Models as well as Generalised Additive Models. In addition, recently critical points in using statistical software for GAM were stressed, and reanalyses of time series data on air pollution and health were performed in order to update already published. Applications are offered through an example on the relationship between asthma emergency admissions and photochemical air pollutants

Natural resource windfalls and efficiency of local government expenditures: evidence from Peru

We study the role of natural resource windfalls in explaining the efficiency of public expenditures. Using a rich dataset of expenditures and public good provision for 1,836 municipalities in Peru for period 2001-2010, we estimate a non-monotonic relationship between the efficiency of public good provision and the level of natural resource transfers. Local governments that were extremely favored by the boom of mineral prices were more efficient in using fiscal windfalls whereas those benefited with modest transfers were more inefficient. These results can be explained by the increase in political competition associated with the boom. However, the fact that increases in efficiency were related to reductions in public good provision casts doubts about the beneficial effects of political competition in promoting efficiency.

Calories, obesity and health in OECD countries

Theoretical models suggest that decisions about diet, weight and health status are endogenous within a utility maximization framework. In this article, we model these behavioural relationships in a fixed-effect panel setting using a simultaneous equation system, with a view to determining whether economic variables can explain the trends in calorie consumption, obesity and health in Organization for Economic Cooperation and Development (OECD) countries and the large differences among the countries. The empirical model shows that progress in medical treatment and health expenditure mitigates mortality from diet-related diseases, despite rising obesity rates. While the model accounts for endogeneity and serial correlation, results are affected by data limitations.

Discrete time representation of continuous time ARMA processes

This paper derives exact discrete time representations for data generated by a continuous time autoregressive moving average (ARMA) system with mixed stock and flow data. The representations for systems comprised entirely of stocks or of flows are also given. In each case the discrete time representations are shown to be of ARMA form, the orders depending on those of the continuous time system. Three examples and applications are also provided, two of which concern the stationary ARMA(2, 1) model with stock variables (with applications to sunspot data and a short-term interest rate) and one concerning the nonstationary ARMA(2, 1) model with a flow variable (with an application to U.S. nondurable consumers’ expenditure). In all three examples the presence of an MA(1) component in the continuous time system has a dramatic impact on eradicating unaccounted-for serial correlation that is present in the discrete time version of the ARMA(2, 0) specification, even though the form of the discrete time model is ARMA(2, 1) for both models.

Climate spectrum estimation in the presence of timescale errors.

We introduce an algorithm (called REDFITmc2) for spectrum estimation in the presence of timescale errors. It is based on the Lomb-Scargle periodogram for unevenly spaced time series, in combination with the Welch's Overlapped Segment Averaging procedure, bootstrap bias correction and persistence estimation. The timescale errors are modelled parametrically and included in the simulations for determining (1) the upper levels of the spectrum of the red-noise AR(1) alternative and (2) the uncertainty of the frequency of a spectral peak. Application of REDFITmc2 to ice core and stalagmite records of palaeoclimate allowed a more realistic evaluation of spectral peaks than when ignoring this source of uncertainty. The results support qualitatively the intuition that stronger effects on the spectrum estimate (decreased detectability and increased frequency uncertainty) occur for higher frequencies. The surplus information brought by algorithm REDFITmc2 is that those effects are quantified. Regarding timescale construction, not only the fixpoints, dating errors and the functional form of the age-depth model play a role. Also the joint distribution of all time points (serial correlation, stratigraphic order) determines spectrum estimation.

Forecasting accuracy and estimation uncertainty using VAR models with short- and long-term economic restrictions: a Monte-Carlo study

Using vector autoregressive (VAR) models and Monte-Carlo simulation methods we investigate the potential gains for forecasting accuracy and estimation uncertainty of two commonly used restrictions arising from economic relationships. The Örst reduces parameter space by imposing long-term restrictions on the behavior of economic variables as discussed by the literature on cointegration, and the second reduces parameter space by imposing short-term restrictions as discussed by the literature on serial-correlation common features (SCCF). Our simulations cover three important issues on model building, estimation, and forecasting. First, we examine the performance of standard and modiÖed information criteria in choosing lag length for cointegrated VARs with SCCF restrictions. Second, we provide a comparison of forecasting accuracy of Ötted VARs when only cointegration restrictions are imposed and when cointegration and SCCF restrictions are jointly imposed. Third, we propose a new estimation algorithm where short- and long-term restrictions interact to estimate the cointegrating and the cofeature spaces respectively. We have three basic results. First, ignoring SCCF restrictions has a high cost in terms of model selection, because standard information criteria chooses too frequently inconsistent models, with too small a lag length. Criteria selecting lag and rank simultaneously have a superior performance in this case. Second, this translates into a superior forecasting performance of the restricted VECM over the VECM, with important improvements in forecasting accuracy ñreaching more than 100% in extreme cases. Third, the new algorithm proposed here fares very well in terms of parameter estimation, even when we consider the estimation of long-term parameters, opening up the discussion of joint estimation of short- and long-term parameters in VAR models.

Estimating the stochastic discount factor without a utility function

Using the Pricing Equation in a panel-data framework, we construct a novel consistent estimator of the stochastic discount factor (SDF) which relies on the fact that its logarithm is the serial-correlation ìcommon featureîin every asset return of the economy. Our estimator is a simple function of asset returns, does not depend on any parametric function representing preferences, is suitable for testing di§erent preference speciÖcations or investigating intertemporal substitution puzzles, and can be a basis to construct an estimator of the risk-free rate. For post-war data, our estimator is close to unity most of the time, yielding an average annual real discount rate of 2.46%. In formal testing, we cannot reject standard preference speciÖcations used in the literature and estimates of the relative risk-aversion coe¢ cient are between 1 and 2, and statistically equal to unity. Using our SDF estimator, we found little signs of the equity-premium puzzle for the U.S.

Ensaios sobre o fator estocástico de descontos

Este trabalho propõe maneiras alternativas para a estimação consistente de uma medida abstrata, crucial para o estudo de decisões intertemporais, o qual é central a grande parte dos estudos em macroeconomia e finanças: o Fator Estocástico de Descontos (SDF, sigla em Inglês). Pelo emprego da Equação de Apreçamento constrói-se um inédito estimador consistente do SDF que depende do fato de que seu logaritmo é comum a todos os ativos de uma economia. O estimador resultante é muito simples de se calcular, não depende de fortes hipóteses econômicas, é adequado ao teste de diversas especificações de preferência e para a investigação de paradoxos de substituição intertemporal, e pode ser usado como base para a construção de um estimador para a taxa livre de risco. Alternativas para a estratégia de identificação são aplicadas e um paralelo entre elas e estratégias de outras metodologias é traçado. Adicionando estrutura ao ambiente inicial, são apresentadas duas situações onde a distribuição assintótica pode ser derivada. Finalmente, as metodologias propostas são aplicadas a conjuntos de dados dos EUA e do Brasil. Especificações de preferência usualmente empregadas na literatura, bem como uma classe de preferências dependentes do estado, são testadas. Os resultados são particularmente interessantes para a economia americana. A aplicação de teste formais não rejeita especificações de preferências comuns na literatura e estimativas para o coeficiente relativo de aversão ao risco se encontram entre 1 e 2, e são estatisticamente indistinguíveis de 1. Adicionalmente, para a classe de preferência s dependentes do estado, trajetórias altamente dinâmicas são estimadas para a tal coeficiente, as trajetórias são confinadas ao intervalo [1,15, 2,05] e se rejeita a hipótese de uma trajetória constante.

What if firms adjust their debt-equity ratios toward a target range?

We estimate optimal target-ranges of capital structure controlling for a series of firmspecific characteristics and accounting for the serial correlation that arises from the dynamic component of the leverage choice. Then, we empirically examine if firms adjust their leverages toward the estimated optimal ranges. Our analysis suggests that the observed behavior of firms is consistent with the notion of range-adjustment.

A Common-feature approach for testing present-value restrictions with financial data

It is well known that cointegration between the level of two variables (labeled Yt and yt in this paper) is a necessary condition to assess the empirical validity of a present-value model (PV and PVM, respectively, hereafter) linking them. The work on cointegration has been so prevalent that it is often overlooked that another necessary condition for the PVM to hold is that the forecast error entailed by the model is orthogonal to the past. The basis of this result is the use of rational expectations in forecasting future values of variables in the PVM. If this condition fails, the present-value equation will not be valid, since it will contain an additional term capturing the (non-zero) conditional expected value of future error terms. Our article has a few novel contributions, but two stand out. First, in testing for PVMs, we advise to split the restrictions implied by PV relationships into orthogonality conditions (or reduced rank restrictions) before additional tests on the value of parameters. We show that PV relationships entail a weak-form common feature relationship as in Hecq, Palm, and Urbain (2006) and in Athanasopoulos, Guillén, Issler and Vahid (2011) and also a polynomial serial-correlation common feature relationship as in Cubadda and Hecq (2001), which represent restrictions on dynamic models which allow several tests for the existence of PV relationships to be used. Because these relationships occur mostly with nancial data, we propose tests based on generalized method of moment (GMM) estimates, where it is straightforward to propose robust tests in the presence of heteroskedasticity. We also propose a robust Wald test developed to investigate the presence of reduced rank models. Their performance is evaluated in a Monte-Carlo exercise. Second, in the context of asset pricing, we propose applying a permanent-transitory (PT) decomposition based on Beveridge and Nelson (1981), which focus on extracting the long-run component of asset prices, a key concept in modern nancial theory as discussed in Alvarez and Jermann (2005), Hansen and Scheinkman (2009), and Nieuwerburgh, Lustig, Verdelhan (2010). Here again we can exploit the results developed in the common cycle literature to easily extract permament and transitory components under both long and also short-run restrictions. The techniques discussed herein are applied to long span annual data on long- and short-term interest rates and on price and dividend for the U.S. economy. In both applications we do not reject the existence of a common cyclical feature vector linking these two series. Extracting the long-run component shows the usefulness of our approach and highlights the presence of asset-pricing bubbles.

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 Araújo 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.

Earnings dynamics and inequality in Venezuela: 1995-1997

In this paper, we decompose the variance of logarithmic monthly earnings of prime age males into its permanent and transitory components, using a five-wave rotating panel from the Venezuelan “Encuesta de Hogares por Muestreo” from 1995 to 1997. As far as we know, this is the first time a variance components model is estimated for a developing country. We test several specifications and find that an error component model with individual random effects and first order serially correlated errors fits the data well. In the simplest model, around 22% of earnings variance is explained by the variance of permanent component, 77% by purely stochastic variation and the remaining 1% by serial correlation. These results contrast with studies from industrial countries where the permanent component is predominant. The permanent component is usually interpreted as the results of productivity characteristics of individuals whereas the transitory component is due to stochastic perturbations such as job and/or price instability, among others. Our findings may be due to the timing of the panel when occurred precisely during macroeconomic turmoil resulting from a severe financial crisis. The findings suggest that earnings instability is an important source of inequality in a region characterized by high inequality and macroeconomic instability.

Double sampling (x)over-bar control chart for a first-order autoregressive moving average process model

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Gráficos de controle de EWMA e de X-barra para monitoramento de processos autocorrelacionados

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)