nep-ecm New Economics Papers
on Econometrics
Issue of 2013‒11‒02
thirteen papers chosen by
Sune Karlsson
Orebro University

  1. Finite-Sample Resampling-Based Combined Hypothesis Tests, with Applications to Serial Correlation and Predictability By Jean-Marie DUFOUR; Lynda KHALAF; Marcel VOIA
  2. Semi-Parametric Inference in Dynamic Binary Choice Models By Andriy Norets; Xun Tang
  3. Testing for Panel Unit Roots under General Cross-Sectional Dependence By Holgersson, Thomas; Månsson, Kristofer; Shukur, Ghazi
  4. On certain transformation of Archimedean copulas: Application to the non-parametric estimation of their generators By Elena Di Bernardino; Didier Rullière
  5. Are Gibbs-type priors the most natural generalization of the Dirichlet process? By Pierpaolo De Blasi; Stefano Favaro; Antonio Lijoi; Ramsés H. Mena; Igor Prünster; Mattteo Ruggiero
  6. Bayesian Variable Selection for Nowcasting Economic Time Series By Steven L. Scott; Hal R. Varian
  7. Forecasting and Nowcasting Macroeconomic Variables: A Methodological Overview By Jennifer Castle; David Hendry
  8. Structural Credit Risk Model with Stochastic Volatility: A Particle-filter Approach By Di Bu; Yin Liao
  9. Understanding and Teaching Within-Cluster Correlation in Complex Surveys By Humberto Barreto; Manu Raghav
  10. Characterizing economic growth paths based on new structural change tests By Nuno Sobreira; Luís Catela Nunes; Paulo M.M. Rodrigues
  11. Modelos Garch Assimétricos com Inovações T-Student By Thais C. O. Fonseca; Vinícios S. Cerqueira; Hélio S. Migon; Cristian A.C.Torres
  12. Modeling and Forecasting Electricity Spot Prices: A Functional Data Perspective By Liebl, Dominik
  13. La régression quantile en pratique By P. GIVORD; X. DHAULTFOEUILLE

  1. By: Jean-Marie DUFOUR; Lynda KHALAF; Marcel VOIA
    Abstract: This paper suggests Monte Carlo multiple test procedures which are provably valid in finite samples. These include combination methods originally proposed for independent statistics and further improvements which formalize statistical practice. We also adapt the Monte Carlo test method to non-continuous combined statistics. The methods suggested are applied to test serial dependence and predictability. In particular, we introduce and analyze new procedures that account for endogenous lag selection. A simulation study illustrates the properties of the proposed methods. Results show that concrete and non-spurious power gains (over standard combination methods) can be achieved through the combined Monte Carlo test approach, and confirm arguments in favour of variance-ratio type criteria.
    Keywords: Monte Carlo test, induced test, test combination, simultaneous inference, Variance ratio
    Date: 2013
  2. By: Andriy Norets (Department of Economics, Princeton University); Xun Tang (Department of Economics, University of Pennsylvania)
    Abstract: We introduce an approach for semi-parametric inference in dynamic binary choice models that does not impose distributional assumptions on the state variables unobserved by the econometrician. The proposed framework combines Bayesian inference with partial identification results. The method is applicable to models with finite space of observed states. We demonstrate the method on Rust's model of bus engine replacement. The estimation experiments show that the parametric assumptions about the distribution of the unobserved states can have a considerable effect on the estimates of per-period payoffs. At the same time, the effect of these assumptions on counterfactual conditional choice probabilities can be small for most of the observed states.
    Keywords: Dynamic discrete choice models, Markov decision processes, semi-parametric inference, identification, Bayesian estimation, MCMC
    JEL: C11 C14
    Date: 2013–10–07
  3. By: Holgersson, Thomas (Jönköping International Business School, and Linnaeus University); Månsson, Kristofer (Jönköping International Business School); Shukur, Ghazi (Jönköping International Business School, and Linnaeus University)
    Abstract: In this paper we generalize four tests of multivariate linear hypothesis to panel data unit root testing. The test statistics are invariant to certain linear transformations of data and therefore simulated critical values may conveniently be used. It is demonstrated that all four tests remains well behaved in cases of where there are heterogeneous alternatives and cross-correlations between marginal variables. A Monte Carlo simulation is included to compare and contrast the tests with two well-established ones.
    Keywords: panel data; unit roots; linear hypothesis; invariance
    JEL: C32 C52
    Date: 2013–10–21
  4. By: Elena Di Bernardino (CEDRIC - Centre d'Etude et De Recherche en Informatique du Cnam - Conservatoire National des Arts et Métiers (CNAM)); Didier Rullière (SAF - Laboratoire de Sciences Actuarielle et Financière - Université Claude Bernard - Lyon I : EA2429)
    Abstract: We study the impact of certain transformations within the class of Archimedean copulas. We give some admissibility conditions for these transformations, and define some equivalence classes for both transformations and generators of Archimedean copulas. We extend the $r$-fold composition of the diagonal section of a copula, from $r \in \N$ to $r \in \R$. This extension, coupled with results on equivalence classes, gives us new expressions of transformations and generators. Estimators deriving directly from these expressions are proposed and their convergence is investigated. We provide confidence bands for the estimated generators. Numerical illustrations show the empirical performance of these estimators.
    Keywords: Transformations of Archimedean copula; self-nested diagonal; non-parametric estimation; tail dependence
    Date: 2013–10–21
  5. By: Pierpaolo De Blasi (Università degli Studi di Torino and Collegio Carlo Alberto); Stefano Favaro (Università degli Studi di Torino and Collegio Carlo Alberto); Antonio Lijoi (Università di Pavia and Collegio Carlo Alberto); Ramsés H. Mena (Universidad Autónoma de México, México); Igor Prünster (Università degli Studi di Torino and Collegio Carlo Alberto); Mattteo Ruggiero (Università degli Studi di Torino and Collegio Carlo Alberto)
    Abstract: Discrete random probability measures and the exchangeable random partitions they induce are key tools for addressing a variety of estimation and prediction problems in Bayesian inference. Indeed, many popular nonparametric priors, such as the Dirichlet and the Pitman–Yor process priors, select discrete probability distributions almost surely and, therefore, automatically induce exchangeable random partitions. Here we focus on the family of Gibbs–type priors, a recent and elegant generalization of the Dirichlet and the Pitman–Yor process priors. These random probability measures share properties that are appealing both from a theoretical and an applied point of view: (i) they admit an intuitive characterization in terms of their predictive structure justifying their use in terms of a precise assumption on the learning mechanism; (ii) they stand out in terms of mathematical tractability; (iii) they include several interesting special cases besides the Dirichlet and the Pitman–Yor processes. The goal of our paper is to provide a systematic and unified treatment of Gibbs–type priors and highlight their implications for Bayesian nonparametric inference. We will deal with their distributional properties, the resulting estimators, frequentist asymptotic validation and the construction of time–dependent versions. Applications, mainly concerning hierarchical mixture models and species sampling, will serve to convey the main ideas. The intuition inherent to this class of priors and the neat results that can be deduced for it lead one to wonder whether it actually represents the most natural generalization of the Dirichlet process.
    Keywords: Bayesian Nonparametrics; Clustering; Consistency; Dependent process; Discrete nonparametric prior; Exchangeable partition probability function; Gibbs–type prior; Pitman–Yor process; Mixture model; Population Genetics; Predictive distribution; Species sampling.
    Date: 2013–10
  6. By: Steven L. Scott; Hal R. Varian
    Abstract: We consider the problem of short-term time series forecasting (nowcasting) when there are more possible predictors than observations. Our approach combines three Bayesian techniques: Kalman filtering, spike-and-slab regression, and model averaging. We illustrate this approach using search engine query data as predictors for consumer sentiment and gun sales.
    JEL: C11 C53
    Date: 2013–10
  7. By: Jennifer Castle; David Hendry
    Abstract: We consider the reasons for nowcasting, how nowcasts can be achieved, and the use and timing of information.� The existence of contemporaneous data such as surveys is a major difference from forecasting, but many of the recent lessons about forecasting remain relevant.� Given the extensive disaggregation over variables underlying flash estimates of aggregates, we show that automatic model selection can play a valuable role, especially when location shifts would otherwise induce nowcast failure.� Thus, we address nowcasting when location shifts occur, probably with measurement error.� We describe impulse-indicator saturation as a potential solution to such shifts, noting its relation to intercept corrections and to robust methods to avoid systematic nowcast failure.� We propose a nowcasting strategy, building models of all disaggregate series by automatic methods, forecasting all variables before the end of each period, testing for shifts as available measures arrive, and adjusting forecasts of cognate missing series if substantive discrepancies are found.� An alternative is switching to robust forecasts when breaks are detected.� We apply a variant of this strategy to nowcast UK GDP growth, seeking pseudo real-time data availability.
    Keywords: Nowcasting, Location shifts, Forecasting, Contemporaneous information, Autometrics, Impulse-indicator saturation
    JEL: C52 C51
    Date: 2013–09–27
  8. By: Di Bu; Yin Liao
    Abstract: This paper extends Merton's structural credit risk model to account for the fact that the firm's asset volatility follows a stochastic process. With the presence of stochastic volatility, the transformed-data maximum likelihood estimation (MLE) method of Duan (1994, 2000) can no longer be applied to estimate the model. We devise a particle filtering algorithm to solve this problem. This algorithm is based on the general non-linear and non-Gaussian filtering with sequential parameter learning, and a simulation study is conducted to ascertain its finite sample performance. Meanwhile, we implement this model on the real data of companies in Dow Jones industrial average and find that incorporating stochastic volatility into the structural model can largely improve the model performance.
    Keywords: Credit risk; Merton model; Stochastic volatility; Particle Filtter; Default probability; CDS
    JEL: C22
    Date: 2013–10–28
  9. By: Humberto Barreto (Department of Economics and Management, DePauw University); Manu Raghav (Department of Economics and Management, DePauw University)
    Abstract: This econometrics pedagogy paper demonstrates the importance of using cluster standard errors with data generated from complex surveys. Simulation is used to show that both classic ordinary least squares and robust standard errors perform poorly in the presence of within-cluster correlated errors, while cluster standard errors are much better. We take advantage of Excel’s spreadsheet interface to produce clear, strong visuals of the data generation process and intuitively explain key results. Stata and R implementations are also provided. We conclude with suggestions for how to use these files in the classroom.
    Keywords: complex survey, simulation, cluster sampling, estimation, survey regression
    JEL: A2 A22 A23 C80 C81 C83 C87
    Date: 2013–07
  10. By: Nuno Sobreira; Luís Catela Nunes; Paulo M.M. Rodrigues
    Abstract: One of the prevalent topics in the economic growth literature is the debate between neoclassical, semi-endogenous, and endogenous growth theories regarding the model that best describes the data. An important part of this discussion can be summarized in three mutually exclusive hypotheses: the \constant trend", the \level shift", and the \slope shift" hypotheses. In this paper we propose the characterization of a country's economic growth path according to these break hypotheses. We address the problem in two steps. First, the number and timing of trend breaks is determined using new structural change tests that are robust to the presence, or not, of unit roots, surpassing technical and methodological concerns of previous empirical studies. Second, conditional on the estimated number of breaks, break dates, and coefficients, a statistical framework is introduced to test for general linear restrictions on the coefficients of the suggested linear disjoint broken trend model. We further show how the aforementioned hypotheses, regarding the economic growth path, can be analysed by a test of linear restrictions on the parameters of the breaking trend model. We apply the methodology to historical per capita GDP for an extensive list of countries. The results support the three alternative hypotheses for different sets of countries.
    JEL: C22 F43 O40
    Date: 2013
  11. By: Thais C. O. Fonseca; Vinícios S. Cerqueira; Hélio S. Migon; Cristian A.C.Torres
    Abstract: Neste trabalho, modela-se a volatilidade usando uma abordagem bayesiana para a estimação de modelos Generalizados Autorregressivos de Heterocedasticidade Condicional – Generalized Autoregressive Conditional Heteroskedasticity (GARCH). Eventuais assimetrias são acomodadas utilizando-se modelos de transição suave para a variância. Apresentam-se alguns problemas relacionados a esta abordagem e discute-se como estes influenciam o comportamento da função de verossimilhança. Para dados cujas distribuições apresentam caudas mais pesadas, utiliza-se a distribuição t-Student. Os problemas da verossimilhança derivados da estimação dos graus de liberdade são resolvidos usando a priori de Jeffrey. Um estudo simulado é apresentado para evidenciar o potencial da metodologia. Por fim, uma aplicação da metodologia a séries de índices de preços ao consumidor (IPCs) no Brasil revela as vantagens da utilização de modelos GARCH assimétricos com distribuição t-Student. In this work we consider modeling the past volatilities through a Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model using the Bayesian approach. Asymmetries in the shocks are accommodated by smooth transition models for the variance. We discuss problems related to the likelihood function and propose a solution. In order to account for heavy tails in the applications we consider Student-terrors. The Jeffrey’s prior is used in this context to correct problems in the estimation of degrees of freedom. A simulated study is presented to highlight the advantages of the proposed methodology and an application to the Brazilian index of prices illustrates the usefulness of the asymmetric GARCH model with student-t errors.
    Date: 2013–09
  12. By: Liebl, Dominik
    Abstract: Classical time series models have serious difficulties in modeling and forecasting the enormous fluctuations of electricity spot prices. Markov regime switch models belong to the most often used models in the electric- ity literature. These models try to capture the fluctuations of electricity spot prices by using different regimes, each with its own mean and covariance structure. Usually one regime is dedicated to moderate prices and another is dedicated to high prices. However, these models show poor performance and there is no theoretical justification for this kind of classification. The merit or- der model, the most important micro-economic pricing model for electricity spot prices, however, suggests a continuum of mean levels with a functional dependence on electricity demand. We propose a new statistical perspective on modeling and forecasting electricity spot prices that accounts for the merit order model. In a first step, the functional relation between electricity spot prices and electricity demand is modeled by daily price-demand functions. In a second step, we parameter- ize the series of daily price-demand functions using a functional factor model. The power of this new perspective is demonstrated by a forecast study that compares our functional factor model with two established classical time se- ries models as well as two alternative functional data models.
    Keywords: Functional factor model, functional data analysis, time series analysis, fundamental market model, merit order curve, European Energy Exchange, EEX
    JEL: C1 C14 C5
    Date: 2013–09
  13. By: P. GIVORD (Insee-Crest); X. DHAULTFOEUILLE (Crest)
    Abstract: Les régressions quantiles sont des outils statistiques dont l'objet est de décrire l'impact de variables explicatives sur une variable d'intérêt. Elles permettent une description plus riche que les régressions linéaires classiques, puisqu'elles s'intéressent à l'ensemble de la distribution conditionnelle de la variable d'intérêt et non seulement à la moyenne de celle-ci. En outre, elles peuvent être plus adaptées pour certains types de données (variables censurées ou tronquées, présence de valeurs extrêmes, modèles non linéaires...). Ce document propose une introduction pratique à ces outils, en insistant sur les détails de leur implémentation pratique par les logiciels statistiques standards (Sas, R, Stata). Il peut également être utilisé comme un guide d'interprétation d'études mobilisant ces méthodes, en s'appuyant sur les deux applications concrètes exposées en détail. Enfin, il présente, pour un public plus averti, des extensions récentes traitant en particulier du traitement de l'endogénéité (variables instrumentales, données de panel...).
    Keywords: Quantile Regression, Quantile Treatment Effect, Instrumental Variable Quantile Regression, Quantile Regression with panel data.
    JEL: C1 C3
    Date: 2013

This nep-ecm issue is ©2013 by Sune Karlsson. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.