Showing papers in "Econometric Reviews in 2016"

Discrete Choice Methods with Simulation

Abstract: Discrete Choice Methods with Simulation by Kenneth Train has been available in the second edition since 2009. The book is published by Cambridge University Press and is also available for download ...

Testing for Serial Correlation in Fixed-Effects Panel Data Models

Abstract: In this paper, we propose various tests for serial correlation in fixed-eects panel data regression models with a small number of time periods. First, a simplified version of the test suggested by Wooldridge (2002) and Drukker (2003) is considered. The second test is based on the Lagrange Multiplier (LM) statistic suggested by Baltagi and Li (1995), and the third test is a modification of the classical Durbin-Watson statistic. Under the null hypothesis of no serial correlation, all tests possess a standard normal limiting distribution as N!1 and T is fixed. Analyzing the local power of the tests, we find that the LM statistic has superior power properties. Furthermore, a generalization to test for autocorrelation up to some given lag order and a test statistic that is robust against time dependent heteroskedasticity are proposed.

Lassoing the HAR Model: A Model Selection Perspective on Realized Volatility Dynamics

Abstract: Realized volatility computed from high-frequency data is an important measure for many applications in finance, and its dynamics have been widely investigated. Recent notable advances that perform well include the heterogeneous autoregressive (HAR) model which can approximate long memory, is very parsimonious, is easy to estimate, and features good out-of-sample performance. We prove that the least absolute shrinkage and selection operator (Lasso) recovers the lags structure of the HAR model asymptotically if it is the true model, and we present Monte Carlo evidence in finite samples. The HAR model's lags structure is not fully in agreement with the one found using the Lasso on real data. Moreover, we provide empirical evidence that there are two clear breaks in structure for most of the assets we consider. These results bring into question the appropriateness of the HAR model for realized volatility. Finally, in an out-of-sample analysis, we show equal performance of the HAR model and the Lasso approach.

Regression Analysis of Multivariate Fractional Data

Abstract: The present article discusses alternative regression models and estimation methods for dealing with multivariate fractional response variables. Both conditional mean models, estimable by nonlinear least squares and quasi-maximum likelihood, and fully parametric models (Dirichlet and Dirichlet-multinomial), estimable by maximum likelihood, are considered. In contrast to previous papers but similarly to the univariate case, a new parameterization is proposed here for the parametric models, which allows the same specification of the conditional mean of interest to be used in all models, irrespective of the specific functional form adopted for it. The text also discusses at some length the specification analysis of fractional regression models, proposing several tests that can be performed through artificial regressions. Finally, an extensive Monte Carlo study evaluates the finite sample properties of most of the estimators and tests considered. JEL classification code: C35.

Common Correlated Effects Estimation of Dynamic Panels with Cross-Sectional Dependence

Abstract: We derive inconsistency expressions for dynamic panel data estimators under error cross-sectional dependence generated by an unobserved common factor in both the fixed effect and the incidental trends case. We show that for a temporally dependent factor, the standard within groups (WG) estimator is inconsistent even as both N and T tend to infinity. Next we investigate the properties of the common correlated effects pooled (CCEP) estimator of Pesaran (2006) which eliminates the error cross-sectional dependence using cross-sectional averages of the data. In contrast to the static case, the CCEP estimator is only consistent when next to N also T tends to infinity. It is shown that for the most relevant parameter settings, the inconsistency of the CCEP estimator is larger than that of the infeasible WG estimator, which includes the common factors as regressors. Restricting the CCEP estimator results in a somewhat smaller inconsistency. The small sample properties of the various estimators are analyzed using ...

Stochastic Model Specification Search for Time-Varying Parameter VARs

Abstract: This article develops a new econometric methodology for performing stochastic model specification search (SMSS) in the vast model space of time-varying parameter vector autoregressions (VARs) with stochastic volatility and correlated state transitions. This is motivated by the concern of overfitting and the typically imprecise inference in these highly parameterized models. For each VAR coefficient, this new method automatically decides whether it is constant or time-varying. Moreover, it can be used to shrink an otherwise unrestricted time-varying parameter VAR to a stationary VAR, thus providing an easy way to (probabilistically) impose stationarity in time-varying parameter models. We demonstrate the effectiveness of the approach with a topical application, where we investigate the dynamic effects of structural shocks in government spending on U.S. taxes and gross domestic product (GDP) during a period of very low interest rates.

Improving the Power of Tests of Stochastic Dominance

Abstract: We extend Hansen's (2005) recentering method to a continuum of inequality constraints to construct new Kolmogorov–Smirnov tests for stochastic dominance of any pre-specified order. We show that our tests have correct size asymptotically, are consistent against fixed alternatives and are unbiased against some N−1/2 local alternatives. It is shown that by avoiding the use of the least favorable configuration, our tests are less conservative and more powerful than Barrett and Donald's (2003) and in some simulation examples we consider, we find that our tests can be more powerful than the subsampling test of Linton et al. (2005). We apply our method to test stochastic dominance relations between Canadian income distributions in 1978 and 1986 as considered in Barrett and Donald (2003) and find that some of the hypothesis testing results are different using the new method.

Inference for Shared-Frailty Survival Models with Left-Truncated Data

Abstract: Shared-frailty survival models specify that systematic unobserved determinants of duration outcomes are identical within groups of individuals. We consider random-effects likelihood-based statistical inference if the duration data are subject to left-truncation. Such inference with left-truncated data can be performed in previous versions of the Stata software package for parametric and semi-parametric shared frailty models. We show that with left-truncated data, the commands ignore the weeding-out process before the left-truncation points, affecting the distribution of unobserved determinants among group members in the data, namely among the group members who survive until their truncation points. We critically examine studies in the statistical literature on this issue as well as published empirical studies that use the commands. Simulations illustrate the size of the (asymptotic) bias and its dependence on the degree of truncation.

The Risk of James–Stein and Lasso Shrinkage

Abstract: This article compares the mean-squared error (or l2 risk) of ordinary least squares (OLS), James–Stein, and least absolute shrinkage and selection operator (Lasso) shrinkage estimators in simple linear regression where the number of regressors is smaller than the sample size. We compare and contrast the known risk bounds for these estimators, which shows that neither James–Stein nor Lasso uniformly dominates the other. We investigate the finite sample risk using a simple simulation experiment. We find that the risk of Lasso estimation is particularly sensitive to coefficient parameterization, and for a significant portion of the parameter space Lasso has higher mean-squared error than OLS. This investigation suggests that there are potential pitfalls arising with Lasso estimation, and simulation studies need to be more attentive to careful exploration of the parameter space.

Modified Profile Likelihood for Fixed-Effects Panel Data Models

Abstract: We show how modified profile likelihood methods, developed in the statistical literature, may be effectively applied to estimate the structural parameters of econometric models for panel data, with a remarkable reduction of bias with respect to ordinary likelihood methods. Initially, the implementation of these methods is illustrated for general models for panel data including individual-specific fixed effects and then, in more detail, for the truncated linear regression model and dynamic regression models for binary data formulated along with different specifications. Simulation studies show the good behavior of the inference based on the modified profile likelihood, even when compared to an ideal, although infeasible, procedure (in which the fixed effects are known) and also to alternative estimators existing in the econometric literature. The proposed estimation methods are implemented in an R package that we make available to the reader.

Robust Ranking of Journal Quality: An Application to Economics

Abstract: The paper focuses on the robustness of rankings of academic journal quality and research impact in general, and in Economics, in particular, based on the widely-used Thomson Reuters ISI Web of Science citations database (ISI). The paper analyses 299 leading international journals in Economics using quantifiable Research Assessment Measures (RAMs), and highlights the similarities and differences in various RAMs, which are based on alternative transformations of citations. All existing RAMs to date have been static, so two new dynamic RAMs are developed to capture changes in impact factor over time and escalating journal self citations. Alternative RAMs may be calculated annually or updated daily to determine When, Where and How (frequently) published papers are cited (see Chang et al. (2011a, b, c)). The RAMs are grouped in four distinct classes that include impact factor, mean citations and non-citations, journal policy, number of high quality papers, and journal influence and article influence. These classes include the most widely used RAMs, namely the classic 2-year impact factor including journal self citations (2YIF), 2-year impact factor excluding journal self citations (2YIF*), 5-year impact factor including journal self citations (5YIF), Eigenfactor (or Journal Influence), Article Influence, h-index, and PI-BETA (Papers Ignored - By Even The Authors). As all existing RAMs to date have been static, two new dynamic RAMs are developed to capture changes in impact factor over time (5YD2 = 5YIF/2YIF) and Escalating Self Citations. We highlight robust rankings based on the harmonic mean of the ranks of RAMs across the 4 classes. It is shown that emphasizing the 2-year impact factor of a journal, which partly answers the question as to When published papers are cited, to the exclusion of other informative RAMs, which answer Where and How (frequently) published papers are cited, can lead to a distorted evaluation of journal quality, impact and influence relative to the harmonic mean of the ranks.

Generalized Least Squares Model Averaging

Abstract: In this article, we propose a method of averaging generalized least squares estimators for linear regression models with heteroskedastic errors. The averaging weights are chosen to minimize Mallows’ Cp-like criterion. We show that the weight vector selected by our method is optimal. It is also shown that this optimality holds even when the variances of the error terms are estimated and the feasible generalized least squares estimators are averaged. The variances can be estimated parametrically or nonparametrically. Monte Carlo simulation results are encouraging. An empirical example illustrates that the proposed method is useful for predicting a measure of firms’ performance.

Has the Relationship between Bank Size and Profitability Changed

Abstract: Kristen Regehr and Rajdeep Sengupta explore whether the relationship between bank size and profitability changed after the 2007–09 financial crisis.

The Local Power of the CADF and CIPS Panel Unit Root Tests

Abstract: Very little is known about the local power of second generation panel unit root tests that are robust to cross-section dependence This article derives the local asymptotic power functions of the cross-section argumented Dickey–Fuller Cross-section Augmented Dickey-Fuller (CADF) and CIPS tests of Pesaran (2007), which are among the most popular tests around

Detection and Estimation of Block Structure in Spatial Weight Matrix

Abstract: In many economic applications, it is often of interest to categorize, classify or label individuals by groups based on similarity of observed behavior. We propose a method that captures group affiliation or, equivalently, estimates the block structure of a neighboring matrix embedded in a Spatial Econometric model. The main results of the LASSO estimator shows that off-diagonal block elements are estimated as zeros with high probability, property defined as “zero-block consistency”. Furthermore, we present and prove zero-block consistency for the estimated spatial weight matrix even under a thin margin of interaction between groups. The tool developed in this paper can be used as a verification of block structure by applied researchers, or as an exploration tool for estimating unknown block structures. We analyzed the US Senate voting data and correctly identified blocks based on party affiliations. Simulations also show that the method performs well.

Conditional VAR and Expected Shortfall: A New Functional Approach

Abstract: We estimate two well-known risk measures, the value-at-risk (VAR) and the expected shortfall, conditionally to a functional variable (i.e., a random variable valued in some semi(pseudo)-metric space). We use nonparametric kernel estimation for constructing estimators of these quantities, under general dependence conditions. Theoretical properties are stated whereas practical aspects are illustrated on simulated data: nonlinear functional and GARCH(1,1) models. Some ideas on bandwidth selection using bootstrap are introduced. Finally, an empirical example is given through data of the S&P 500 time series.

Conditionally Efficient Estimation of Long-Run Relationships Using Mixed-Frequency Time Series

Abstract: I analyze efficient estimation of a cointegrating vector when the regressand and regressor are observed at different frequencies. Previous authors have examined the effects of specific temporal aggregation or sampling schemes, finding conventionally efficient techniques to be efficient only when both the regressand and the regressors are average sampled. Using an alternative method for analyzing aggregation under more general weighting schemes, I derive an efficiency bound that is conditional on the type of aggregation used on the low-frequency series and differs from the unconditional bound defined by the full-information high-frequency data-generating process, which is infeasible due to aggregation of at least one series. I modify a conventional estimator, canonical cointegrating regression (CCR), to accommodate cases in which the aggregation weights are known. The correlation structure may be utilized to offset the potential information loss from aggregation, resulting in a conditionally efficient esti...

The Performance of Lag Selection and Detrending Methods for HEGY Seasonal Unit Root Tests

Abstract: This paper analyzes two key issues for the empirical implementation of parametric seasonal unit root tests, namely generalized least squares (GLS) versus ordinary least squares (OLS) detrending and the selection of the lag augmentation polynomial. Through an extensive Monte Carlo analysis, the performance of a battery of lag selection techniques is analyzed, including a new extension of modified information criteria for the seasonal unit root context. All procedures are applied for both OLS and GLS detrending for a range of data generating processes, also including an examination of hybrid OLS-GLS detrending in conjunction with (seasonal) modified AIC lag selection. An application to quarterly U.S. industrial production indices illustrates the practical implications of choices made.

The structural causes of Japan’s Lost Decades

Abstract: Although by the early 2000s, Japan had largely overcome its non-performing loan problems, economic growth hardly accelerated, resulting in what now are "two lost decades." This paper examines the underlying reasons from a long-term and structural perspective. Major issues examined include the chronic lack of domestic demand since the mid-1970s, caused by a long-run decline in capital formation through a slowdown in the growth of the working age population and resulting in a current account surplus and yen appreciation, and supply-side issues such as stagnant TFP growth caused by Japan's low economic metabolism. A key finding is that whereas large firms since the mid-1990s have achieved greater increases in TFP than in the 1980s through R&D and internationalization, the TFP of small firms has stagnated.

On the Usefulness or Lack Thereof of Optimality Criteria for Structural Change Tests

Abstract: Elliott and Muller (2006) considered the problem of testing for general types of parameter variations, including infrequent breaks. They developed a framework that yields optimal tests, in the sense that they nearly attain some local Gaussian power envelop. The main ingredient in their setup is that the variance of the process generating the changes in the parameters must go to zero at a fast rate. They recommended the so-called qLL test, a partial sums type test based on the residuals obtained from the restricted model. We show that for breaks that are very small, its power is indeed higher than other tests, including the popular sup-Wald (SW) test. However, the differences are very minor. When the magnitude of change is moderate to large, the power of the test is very low in the context of a regression with lagged dependent variables or when a correction is applied to account for serial correlation in the errors. In many cases, the power goes to zero as the magnitude of change increases. The power of t...

A Goodness-of-Fit Test for a Class of Autoregressive Conditional Duration Models

Abstract: This article develops a method for testing the goodness-of-fit of a given parametric autoregressive conditional duration model against unspecified nonparametric alternatives. The test statistics are functions of the residuals corresponding to the quasi maximum likelihood estimate of the given parametric model, and are easy to compute. The limiting distributions of the test statistics are not free from nuisance parameters. Hence, critical values cannot be tabulated for general use. A bootstrap procedure is proposed to implement the tests, and its asymptotic validity is established. The finite sample performances of the proposed tests and several other competing ones in the literature, were compared using a simulation study. The tests proposed in this article performed well consistently throughout, and they were either the best or close to the best. None of the tests performed uniformly the best. The tests are illustrated using an empirical example.

Heteroskedasticity Robust Panel Unit Root Testing Under Variance Breaks in Pooled Regressions

Abstract: Noting that many economic variables display occasional shifts in their second order moments, we investigate the performance of homogenous panel unit root tests in the presence of permanent volatility shifts. It is shown that in this case the test statistic proposed by Herwartz and Siedenburg (2008) is asymptotically standard Gaussian. By means of a simulation study we illustrate the performance of first and second generation panel unit root tests and undertake a more detailed comparison of the test in Herwartz and Siedenburg (2008) and its heteroskedasticity consistent Cauchy counterpart introduced in Demetrescu and Hanck (2012a). As an empirical illustration, we reassess evidence on the Fisher hypothesis with data from nine countries over the period 1961Q2–2011Q2. Empirical evidence supports panel stationarity of the real interest rate for the entire subperiod. With regard to the most recent two decades, the test results cast doubts on market integration, since the real interest rate is diagnosed nonstat...

Random Effects, Fixed Effects and Hausman's Test for the Generalized Mixed Regressive Spatial Autoregressive Panel Data Model

Abstract: This article suggests random and fixed effects spatial two-stage least squares estimators for the generalized mixed regressive spatial autoregressive panel data model. This extends the generalized spatial panel model of Baltagi et al. (2013) by the inclusion of a spatial lag term. The estimation method utilizes the Generalized Moments method suggested by Kapoor et al. (2007) for a spatial autoregressive panel data model. We derive the asymptotic distributions of these estimators and suggest a Hausman test a la Mutl and Pfaffermayr (2011) based on the difference between these estimators. Monte Carlo experiments are performed to investigate the performance of these estimators as well as the corresponding Hausman test.

Forecasting Macroeconomic Variables Using Neural Network Models and Three Automated Model Selection Techniques

Abstract: When forecasting with neural network models one faces several problems, all of which influence the accuracy of the forecasts. First, neural networks are often hard to estimate due to their highly nonlinear structure. To alleviate the problem, White (2006) presented a solution (QuickNet) that converts the specification and nonlinear estimation problem into a linear model selection and estimation problem. We shall compare its performance to that of two other procedures building on the linearization idea: the Marginal Bridge Estimator and Autometrics. Second, one must decide whether forecasting should be carried out recursively or directly. This choice is investigated in this work. The economic time series used in this study are the consumer price indices for the G7 and the Scandinavian countries. In addition, a number of simulations are carried out and results reported in the article.

Efficient Estimation with Many Weak Instruments Using Regularization Techniques

Abstract: The problem of weak instruments is due to a very small concentration parameter. To boost the concentration parameter, we propose to increase the number of instruments to a large number or even up to a continuum. However, in finite samples, the inclusion of an excessive number of moments may be harmful. To address this issue, we use regularization techniques as in Carrasco (2012) and Carrasco and Tchuente (2014). We show that normalized regularized two-stage least squares (2SLS) and limited maximum likelihood (LIML) are consistent and asymptotically normally distributed. Moreover, our estimators are asymptotically more efficient than most competing estimators. Our simulations show that the leading regularized estimators (LF and T of LIML) work very well (are nearly median unbiased) even in the case of relatively weak instruments. An application to the effect of institutions on output growth completes the article.

Monetary Policy at the Zero Lower Bound: Revelations from the FOMC's Summary of Economic Projections

Abstract: George A. Kahn and Andrew Palmer assess how FOMC participants' projections that policy would lift off from its effective lower bound related to their projections for inflation and unemployment. The article is summarized in The Macro Bulletin.

The Lasting Damage from the Financial Crisis to U.S. Productivity

Abstract: Michael Redmond and Willem Van Zandweghe find that tight credit conditions during the 2007–09 financial crisis dampened productivity, leaving it on a lower trajectory. The article is summarized in The Macro Bulletin.

Inference of the Trend in a Partially Linear Model with Locally Stationary Regressors

Abstract: In this article, we construct the uniform confidence band (UCB) of nonparametric trend in a partially linear model with locally stationary regressors. A two-stage semiparametric regression is employed to estimate the trend function. Based on this estimate, we develop an invariance principle to construct the UCB of the trend function. The proposed methodology is used to estimate the Non-Accelerating Inflation Rate of Unemployment (NAIRU) in the Phillips Curve and to perform inference of the parameter based on its UCB. The empirical results strongly suggest that the U.S. NAIRU is time-varying.

Particle Learning for Fat-Tailed Distributions

Abstract: It is well known that parameter estimates and forecasts are sensitive to assumptions about the tail behavior of the error distribution. In this article, we develop an approach to sequential inference that also simultaneously estimates the tail of the accompanying error distribution. Our simulation-based approach models errors with a tν-distribution and, as new data arrives, we sequentially compute the marginal posterior distribution of the tail thickness. Our method naturally incorporates fat-tailed error distributions and can be extended to other data features such as stochastic volatility. We show that the sequential Bayes factor provides an optimal test of fat-tails versus normality. We provide an empirical and theoretical analysis of the rate of learning of tail thickness under a default Jeffreys prior. We illustrate our sequential methodology on the British pound/U.S. dollar daily exchange rate data and on data from the 2008–2009 credit crisis using daily S&P500 returns. Our method naturally extends ...

Lassoing the determinants of retirement

Abstract: This article uses Danish register data to explain the retirement decision of workers in 1990 and 1998. Many variables might be conjectured to influence this decision such as demographic, socioeconomic, financial, and health related variables as well as all the same factors for the spouse in case the individual is married. In total, we have access to 399 individual specific variables that all could potentially impact the retirement decision. We use variants of the least absolute shrinkage and selection operator (Lasso) and the adaptive Lasso applied to logistic regression in order to uncover determinants of the retirement decision. To the best of our knowledge, this is the first application of these estimators in microeconometrics to a problem of this type and scale. Furthermore, we investigate whether the factors influencing the retirement decision are stable over time, gender, and marital status. It is found that this is the case for core variables such as age, income, wealth, and general health. We also...