
on Econometrics 
By:  Nese Yildiz (University of Rochester, Department of Economics) 
Abstract:  This paper presents computationally simple estimators for the index coefficients in a binary choice model with a binary endogenous regressor without relying on distributional assumptions or on large support conditions and yields rootn consistent and asymptotically normal estimators. We develop a multistep method for estimating the parameters in a triangular, linear index, thresholdcrossing model with two equations. Such an econometric model might be used in testing for moral hazard while allowing for asymmetric information in insurance markets. In outlining this new estimation method two contributions are made. The first one is proposing a novel ”matching” estimator for the coefficient on the binary endogenous variable in the outcome equation. Second, in order to establish the asymptotic properties of the proposed estimators for the coefficients of the exogenous regressors in the outcome equation, the results of Powell, Stock and Stoker (1989) are extended to cover the case where the average derivative estimation requires a first step semiparametric procedure. 
Date:  2012–01 
URL:  http://d.repec.org/n?u=RePEc:koc:wpaper:1202&r=ecm 
By:  Lavergne, Pascal; Patilea, Valentin 
Abstract:  We develop a novel approach to build checks of parametric regression models when many regressors are present, based on a class of sufficiently rich semiparametric alternatives, namely singleindex models. We propose an omnibus test based on the kernel method that performs against a sequence of directional nonparametric alternatives as if there was one regressor only, whatever the number of regressors. This test can be viewed as a smooth version of the integrated conditional moment (ICM) test of Bierens. Qualitative information can be easily incorporated into the procedure to enhance power. In an extensive comparative simulation study, we find that our test is little sensitive to the smoothing parameter and performs well in multidimensional settings. We then apply it to a crosscountry growth regression model. 
Keywords:  Dimensionality; Hypothesis testing; Nonparametric methods 
JEL:  C14 C12 
Date:  2011 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:35779&r=ecm 
By:  Peter C.B. Phillips (Cowles Foundation, Yale University); ShuPing Shi (Australian National University); Jun Yu (Singapore Management University) 
Abstract:  Righttailed unit root tests have proved promising for detecting exuberance in economic and financial activities. Like lefttailed tests, the limit theory and test performance are sensitive to the null hypothesis and the model specification used in parameter estimation. This paper aims to provide some empirical guidelines for the practical implementation of righttailed unit root tests, focussing on the sup ADF test of Phillips, Wu and Yu (2011), which implements a righttailed ADF test repeatedly on a sequence of forward sample recursions. We analyze and compare the limit theory of the sup ADF test under different hypotheses and model specifications. The size and power properties of the test under various scenarios are examined in simulations and some recommendations for empirical practice are given. Empirical applications to the Nasdaq and to Australian and New Zealand housing data illustrate these specification issues and reveal their practical importance in testing. 
Keywords:  Unit root test, Mildly explosive process, Recursive regression, Size and power 
JEL:  C15 C22 
Date:  2012–01 
URL:  http://d.repec.org/n?u=RePEc:cwl:cwldpp:1842&r=ecm 
By:  R. ALHAMZAWI; K. YU; D. F. BENOIT 
Abstract:  Recently, variable selection by penalized likelihood has attracted much research interest. In this paper, we propose adaptive Lasso quantile regression (BALQR) from a Bayesian perspective. The method extends the Bayesian Lasso quantile regression by allowing different penalization parameters for different regression coefficients. Inverse gamma prior distributions are placed on the penalty parameters. We treat the hyperparameters of the inverse gamma prior as unknowns and estimate them along with the other parameters. A Gibbs sampler is developed to simulate the parameters from the posterior distributions. Through simulation studies and analysis of a prostate cancer data set, we compare the performance of the BALQR method proposed with six existing Bayesian and nonBayesian methods. The simulation studies and the prostate cancer data analysis indicate that the BALQR method performs well in comparision to the other approaches. 
Keywords:  Gibbs sampler, Lasso, Quantile regression, Skewed Laplace distribution. 
Date:  2011–07 
URL:  http://d.repec.org/n?u=RePEc:rug:rugwps:11/728&r=ecm 
By:  Yuanhua Feng (University of Paderborn) 
Abstract:  This paper proposes a local linear estimator for diurnal patterns of transaction durations under a special nonparametric regression model, whose asymptotics are different to any known results. An iterative plugin algorithm is developed for selecting the bandwidth. The ACD model is then applied to analyze the standardized durations. Data examples show that the proposals work well in practice. 
Keywords:  Autoregressive conditional duration, diurnal duration patterns, local linear estimator, bandwidth selection, iterative plugin. 
JEL:  C14 C41 
Date:  2011–12 
URL:  http://d.repec.org/n?u=RePEc:pdn:wpaper:44&r=ecm 
By:  Muni S. Srivastava (Department of Statistics, University of Toronto); Tatsuya Kubokawa (Faculty of Economics, University of Tokyo) 
Abstract:  In this article, we consider the problem of testing the equality of mean vectors of dimension ρ of several groups with a common unknown nonsingular covariance matrix Σ, based on <em>N</em> independent observation vectors where <em>N</em> may be less than the dimension ρ. This problem, known in the literature as the Multivariate Analysis of variance (MANOVA) in highdimension has recently been considered in the statistical literature by Srivastava and Fujikoshi[7], Srivastava [5] and Schott[3]. All these tests are not invariant under the change of units of measurements. On the lines of Srivastava and Du[8] and Srivastava[6], we propose a test that has the above invariance property. The null and the nonnull distributions are derived under the assumption that (<em>N</em>, ρ) → ∞ and <em>N</em> may be less than ρ and the observation vectors follow a general nonnormal model. 
Date:  2011–12 
URL:  http://d.repec.org/n?u=RePEc:tky:fseres:2011cf831&r=ecm 
By:  Matei Demetrescu (University of Bonn); Robinson Kruse (Leibniz University Hannover and CREATES) 
Abstract:  This article extends the analysis of local power of unit root tests in a nonlinear direction by considering local nonlinear alternatives and tests built specically against stationary nonlinear models. In particular, we focus on the popular test proposed by Kapetanios et al. (2003, Journal of Econometrics 112, 359379) in comparison to the linear DickeyFuller test. To this end, we consider different adjustment schemes for deterministic terms. We provide asymptotic results which imply that the error variance has a severe impact on the behavior of the tests in the nonlinear case; the reason for such behavior is the interplay of nonstationarity and nonlinearity. In particular, we show that nonlinearity of the data generating process can be asymptotically negligible when the error variance is moderate or large (compared to the "amount of nonlinearity"), rendering the linear test more powerful than the nonlinear one. Should however the error variance be small, the nonlinear test has better power against local alternatives. We illustrate this in an asymptotic framework of what we call persistent nonlinearity. The theoretical findings of this article explain previous results in the literature obtained by simulation. Furthermore, our own simulation results suggest that the userspecied adjustment scheme for deterministic components (e.g. OLS, GLS, or recursive adjustment) has a much higher impact on the power of unit root tests than accounting for nonlinearity, at least under local (linear or nonlinear) alternatives. 
Keywords:  Nonlinear models, Stochastic trend, Near integration, Persistent nonlinearity, Local power 
JEL:  C12 C22 
Date:  2012–01–04 
URL:  http://d.repec.org/n?u=RePEc:aah:create:201201&r=ecm 
By:  Peter C.B. Phillips (Cowles Foundation, Yale University); Ji Hyung Lee (Dept. of Economics, Yale University) 
Abstract:  Limit theory is developed for nonstationary vector autoregression (VAR) with mixed roots in the vicinity of unity involving persistent and explosive components. Statistical tests for common roots are examined and model selection approaches for discriminating roots are explored. The results are useful in empirical testing for multiple manifestations of nonstationarity  in particular for distinguishing mildly explosive roots from roots that are local to unity and for testing commonality in persistence. 
Keywords:  Common roots, Local to unity, Mildly explosive, Mixed roots, Model selection, Persistence, Tests of common roots 
JEL:  C22 
Date:  2012–01 
URL:  http://d.repec.org/n?u=RePEc:cwl:cwldpp:1845&r=ecm 
By:  Yevgeniy Kovchegov (University of Rochester, Department of Mathematics); Nese Yildiz (Oregon State University, Department of Economics) 
Abstract:  We solve a class of identification problems for nonparametric and semiparametric models when the endogenous covariate is discrete with unbounded support. Then we proceed with an approach that resolves a polynomial basis problem for the above class of discrete distributions, and for the distributions given in the sufficient condition for completeness in Newey and Powell (2003). Thus, in addition to extending the set of econometric models for which nonparametric or semiparametric identification of structural functions is guaranteed to hold, our approach provides a natural way of estimating these functions. Finally, we extend our polynomial basis approach to Pearsonlike and Ordlike families of distributions. 
Keywords:  nonparametric methods, identification, instrumental variables. 
Date:  2012–01 
URL:  http://d.repec.org/n?u=RePEc:koc:wpaper:1203&r=ecm 
By:  Alastair R. Hall; Yuyi Li; Chris D. Orme 
Date:  2012 
URL:  http://d.repec.org/n?u=RePEc:man:sespap:1205&r=ecm 
By:  M. FRÖMMEL; R. KRUSE 
Abstract:  We analyze the time series properties of the S&P500 dividendprice ratio in the light of long memory, structural breaks and rational bubbles. We find an increase in the long memory parameter in the early 1990s by applying a recently proposed test by Sibbertsen and Kruse (2009). An application of the unit root test against long memory by Demetrescu et al. (2008) suggests that the prebreak data can be characterized by long memory, while the postbreak sample contains a unit root. These results reconcile two empirical findings which were seen as contradictory so far: on the one hand they confirm the existence of fractional integration in the S&P500 log dividendprice ratio and on the other hand they are consistent with the existence of a rational bubble. The result of a changing memory parameter in the dividendprice ratio has an important implication for the literature on return predictability: the shift from a stationary dividendprice ratio to a unit root process in 1991 is likely to have caused the welldocumented failure of conventional return prediction models since the 1990s. 
Keywords:  Rational bubbles, dividendprice ratio, fractional integration, changing persistence. 
JEL:  C12 C22 G12 
Date:  2011–05 
URL:  http://d.repec.org/n?u=RePEc:rug:rugwps:11/722&r=ecm 
By:  Anindya Banerjee; Josep Lluis CarrioniSilvestre 
Abstract:  Panel cointegration, structural break, common factors, crosssection dependence 
Keywords:  Panel cointegration, structural break, common factors, crosssection dependence 
JEL:  C12 C22 
Date:  2011–12 
URL:  http://d.repec.org/n?u=RePEc:bir:birmec:1125&r=ecm 
By:  Richard Nickl; Markus Reiß 
Abstract:  Given n equidistant realisations of a Lévy process (Lt; t >= 0), a natural estimator for the distribution function N of the Lévy measure is constructed. Under a polynomial decay restriction on the characteristic function, a Donskertype theorem is proved, that is, a functional central limit theorem for the process in the space of bounded functions away from zero. The limit distribution is a generalised Brownian bridge process with bounded and continuous sample paths whose covariance structure depends on the Fourierintegral operator. The class of Lévy processes covered includes several relevant examples such as compound Poisson, Gamma and selfdecomposable processes. Main ideas in the proof include establishing pseudolocality of the Fourierintegral operator and recent techniques from smoothed empirical processes. 
Keywords:  uniform central limit theorem, nonlinear inverse problem, smoothed empirical processes, pseudodifferential operators, jump measure 
JEL:  C14 C22 
Date:  2012–01 
URL:  http://d.repec.org/n?u=RePEc:hum:wpaper:sfb649dp2012003&r=ecm 
By:  Fred Espen Benth; Claudia Kl\"uppelberg; Gernot M\"uller; Linda Vos 
Abstract:  We present a new model for the electricity spot price dynamics, which is able to capture seasonality, lowfrequency dynamics and the extreme spikes in the market. Instead of the usual purely deterministic trend we introduce a nonstationary independent increments process for the lowfrequency dynamics, and model the large fluctuations by a nonGaussian stable CARMA process. The model allows for analytic futures prices, and we apply these to model and estimate the whole market consistently. Besides standard parameter estimation, an estimation procedure is suggested, where we fit the nonstationary trend using futures data with long time until delivery, and a robust $L^1$filter to find the states of the CARMA process. The procedure also involves the empirical and theoretical risk premiums which  as a byproduct  are also estimated. We apply this procedure to data from the German electricity exchange EEX, where we split the empirical analysis into base load and peak load prices. We find an overall negative risk premium for the base load futures contracts, except for contracts close to delivery, where a small positive risk premium is detected. The peak load contracts, on the other hand, show a clear positive risk premium, when they are close to delivery, while the contracts in the longer end also have a negative premium. 
Date:  2012–01 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:1201.1151&r=ecm 
By:  Andras Fulop (Finance Department, ESSEC Business School, ParisSingapore, CergyPontoise Cedex, France 95021); Junye Li (Finance Department, ESSEC Business School, ParisSingapore, 100 Victoria Street, Singapore 188064); Jun Yu (Sim Kee Boon Institute for Financial Economics, School of Economics, and Lee Kong Chian School of Business, Singapore Management University, 90 Stamford Road, Singapore 178903) 
Abstract:  The paper proposes a new class of continuoustime asset pricing models where negative jumps play a crucial role. Whenever there is a negative jump in asset returns, it is simultaneously passed on to diffusion variance and the jump intensity, generating selfexciting cojumps of prices and volatility and jump clustering. To properly deal with parameter uncertainty and insample overfitting, a Bayesian learning approach combined with an efficient particle filter is employed. It not only allows for comparison of both nested and nonnested models, but also generates all quantities necessary for sequential model analysis. Empirical investigation using S&P 500 index returns shows that volatility jumps at the same time as negative jumps in asset returns mainly through jumps in diffusion volatility. We find substantial evidence for jump clustering, in particular, after the recent financial crisis in 2008, even though parameters driving dynamics of the jump intensity remain difficult to identify. 
Keywords:  SelfExcitation, Volatility Jump, Jump Clustering, Extreme Events, Parameter Learning, Particle Filters, Sequential Bayes Factor, Risk Management 
JEL:  C11 C13 C32 G12 
Date:  2012–01 
URL:  http://d.repec.org/n?u=RePEc:siu:wpaper:032012&r=ecm 