
on Econometrics 
By:  Linton, O.; Seo, M.; Whang, YJ. 
Abstract:  We propose a test of the hypothesis of conditional stochastic dominance in the presence of many conditioning variables (whose dimension may grow to infinity as the sample size diverges). Our approach builds on a semiparametric location scale model in the sense that the conditional distribution of the outcome given the covariates is characterized by a nonparametric mean function and a nonparametric skedastic function with an independent innovation whose distribution is unknown. We propose to estimate the nonparametric mean and skedastic regression functions by the `1penalized nonparametric series estimation with thresholding. Under the sparsity assumption, where the number of truly relevant series terms are relatively small (but their identities are unknown), we develop the estimation error bounds for the regression functions and series coefficients estimates allowing for the time series dependence. We derive the asymptotic distribution of the test statistic, which is not pivotal asymptotically, and introduce the smooth stationary bootstrap to approximate its sample distribution. We investigate the finite sample performance of the bootstrap critical values by a set of Monte Carlo simulations. Finally, our method is illustrated by an application to stochastic dominance among portfolio returns given all the past information. 
Keywords:  Bootstrap, Empirical process, Home bias, LASSO, Power boosting, Sparsity 
JEL:  C10 C12 C15 C15 
Date:  2020–01–14 
URL:  http://d.repec.org/n?u=RePEc:cam:camdae:2004&r=all 
By:  Wu, R.; Weeks, M. 
Abstract:  Generalized Least Square (GLS) estimators have been vastly applied in empirical studies to improve the efficiency of estimation. However, parametric GLS still imposes certain assumptions on the form of the covariance matrix of the unobservable, and the efficiency gain of GLS in fact depends on these assumptions being correct. In this paper we propose a semiparametric Bayesian GLS estimator to cope with such heterogeneity. A Dirichlet process prior is put on the distribution of the covariance matrices of the unobservables, leading to a model that could be interpreted as the mixture of a variable number of normal distributions. Our methods let the number of heterogeneous groups be data driven, and so is the group membership of each observation. The semiparametric Bayesian Seemingly Unrelated Regression (SUR) for equation systems, as well as Random Effects Model (REM) and Correlated Random Effects Model (CREM) for panel data are then described as special cases of the GLS estimators. A series of simulation experiments is designed to explore the performance of our methods, and demonstrates that they provide more reliable inference than the parametric Bayesian GLS. We then apply our semiparametric Bayesian SUR and REM/CREM methods to empirical examples. 
Keywords:  Bayesian semiparametric, generalized lease square estimator, Dirichlet process, equation system, seemingly unrelated regression, panel data, random effects model, correlated random effects model. 
JEL:  C30 
Date:  2020–02–20 
URL:  http://d.repec.org/n?u=RePEc:cam:camdae:2011&r=all 
By:  Jiaming Mao; Jingzhi Xu 
Abstract:  Statistical and structural modeling represent two distinct approaches to data analysis. In this paper, we propose a set of novel methods for combining statistical and structural models for improved prediction and causal inference. Our first proposed estimator has the doubly robustness property in that it only requires the correct specification of either the statistical or the structural model. Our second proposed estimator is a weighted ensemble that has the ability to outperform both models when they are both misspecified. Experiments demonstrate the potential of our estimators in various settings, including fistprice auctions, dynamic models of entry and exit, and demand estimation with instrumental variables. 
Date:  2020–06 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:2006.05308&r=all 
By:  William C. Horrace (Center for Policy Research, Maxwell School, Syracuse University, 426 Eggers Hall, Syracuse, NY 13244); Yulong Wang (Center for Policy Research, Maxwell School, Syracuse University, 426 Eggers Hall, Syracuse, NY 13244) 
Abstract:  This article studies tail behavior for the error components in the stochastic frontier model, where one component has bounded support on one side, and the other has unbounded support on both sides. Under weak assumptions on the error components, we derive nonparametric tests that the unbounded component distribution has thin tails and that the component tails are equivalent. The tests are useful diagnostic tools for stochastic frontier analysis and kernel deconvolution density estimation. A simulation study and an application to a stochastic cost frontier for 6,100 US banks from 1998 to 2005 are provided. The new tests reject the normal or Laplace distributional assumptions, which are commonly imposed in the existing literature. 
Keywords:  Hypothesis Testing, Production, Inefficiency, Deconvolution, Extreme Value Theory 
JEL:  C12 C21 D24 
Date:  2020–06 
URL:  http://d.repec.org/n?u=RePEc:max:cprwps:230&r=all 
By:  Dendramis, Yiannis; kapetanios, george; Marcellino, Massimiliano 
Abstract:  In the aftermath of the recent financial crisis there has been considerable focus on methods for predicting macroeconomic variables when their behavior is subject to abrupt changes, associated for example with crisis periods. In this paper we propose similarity based approaches as a way to handle parameter instability, and apply them to macroeconomic forecasting. The rationale is that clusters of past data that match the current economic conditions can be more informative for forecasting than the entire past behavior of the variable of interest. We apply our methods to predict both simulated data in a set of Monte Carlo experiments, and a broad set of key US macroeconomic indicators. The forecast evaluation exercises indicate that similaritybased approaches perform well, in general, in comparison with other common timevarying forecasting methods, and particularly well during crisis episodes. 
Keywords:  empirical similarity; Forecast comparison; Kernel estimation ; Macroeconomic forecasting; parameter time variation 
Date:  2020–03 
URL:  http://d.repec.org/n?u=RePEc:cpr:ceprdp:14469&r=all 
By:  Onatski, A.; Wang, C. 
Abstract:  This paper draws parallels between the Principal Components Analysis of factorless highdimensional nonstationary data and the classical spurious regression. We show that a few of the principal components of such data absorb nearly all the data variation. The corresponding scree plot suggests that the data contain a few factors, which is collaborated by the standard panel information criteria. Furthermore, the DickeyFuller tests of the unit root hypothesis applied to the estimated “idiosyncratic terms” often reject, creating an impression that a few factors are responsible for most of the nonstationarity in the data. We warn empirical researchers of these peculiar effects and suggest to always compare the analysis in levels with that in differences. 
Keywords:  Spurious regression, principal components, factor models, KarhunenLoève expansion. 
Date:  2020–01–13 
URL:  http://d.repec.org/n?u=RePEc:cam:camdae:2003&r=all 
By:  Kelly, Morgan 
Abstract:  Abstract Regressions using data with known locations are increasingly used in empirical economics, and several standard error corrections are available to deal with the fact that their residuals tend to be spatially correlated. Unfortunately, different corrections commonly return significance levels that vary by several orders of magnitude, leaving the researcher uncertain as to which, if any, is valid. This paper proposes instead an extremely fast and simple procedure to derive standard errors directly from the spatial correlation structure of regression residuals. Importantly, because the estimated covariance matrix gives optimal weights to predict each residual as a linear combination of all residuals, the reliability of these standard errors is selfchecking by construction. The approach extends immediately to instrumental variables, balanced and unbalanced panels, and a wide class of nonlinear models. A step by step guide to estimating these standard errors is given in the accompanying tutorials. 
Date:  2020–03 
URL:  http://d.repec.org/n?u=RePEc:cpr:ceprdp:14483&r=all 
By:  Kilian, Lutz; Zhou, Xiaoqing 
Abstract:  Oil market VAR models have become the standard tool for understanding the evolution of the real price of oil and its impact in the macro economy. As this literature has expanded at a rapid pace, it has become increasingly difficult for mainstream economists to understand the differences between alternative oil market models, let alone the basis for the sometimes divergent conclusions reached in the literature. The purpose of this survey is to provide a guide to this literature. Our focus is on the econometric foundations of the analysis of oil market models with special attention to the identifying assumptions and methods of inference. We not only explain how the workhorse models in this literature have evolved, but also examine alternative oil market VAR models. We help the reader understand why the latter models sometimes generated unconventional, puzzling or erroneous conclusions. Finally, we discuss the construction of extraneous measures of oil demand and oil supply shocks that have been used as external or internal instruments for VAR models. 
Keywords:  Bayesian estimation; Elasticity; identification; Model specification; structural VAR; textual analysis 
JEL:  C36 C52 Q41 Q43 
Date:  2020–03 
URL:  http://d.repec.org/n?u=RePEc:cpr:ceprdp:14460&r=all 
By:  Ek, Claes (Department of Economics, School of Business, Economics and Law, Göteborg University) 
Abstract:  Recent research by Burlig et al. (2020) has produced a useful formula for performing differenceindifferences power calculation in the presence of serially correlated errors. A similar formula for the ANCOVA estimator is shown by the authors to yield incorrect power in real data where time shocks are present. This note demonstrates that the serialcorrelationrobust ANCOVA formula is in fact correct under time shocks as well. The reason that errors arise in Burlig et al. (2020) is because time shocks remain unaccounted for in the intermediate step where residualbased variance parameters are estimated from preexisting data. When that procedure is adjusted accordingly, the serialcorrelationrobust ANCOVA formula of Burlig et al. (2020) can be accurately used for power calculation. 
Keywords:  power calculation; randomized experiments; experimental design; panel data; ANCOVA 
JEL:  C23 C93 
Date:  2020–06 
URL:  http://d.repec.org/n?u=RePEc:hhs:gunwpe:0788&r=all 
By:  Christoph Berninger; Almond St\"ocker; David R\"ugamer 
Abstract:  Motivated by the application to German interest rates, we propose a timevarying autoregressive model for short and long term prediction of time series that exhibit a temporary nonstationary behavior but are assumed to mean revert in the long run. We use a Bayesian formulation to incorporate prior assumptions on the mean reverting process in the model and thereby regularize predictions in the far future. We use MCMCbased inference by deriving relevant full conditional distributions and employ a MetropolisHastings within Gibbs Sampler approach to sample from the posterior (predictive) distribution. In combining datadriven short term predictions with long term distribution assumptions our model is competitive to the existing methods in the short horizon while yielding reasonable predictions in the long run. We apply our model to interest rate data and contrast the forecasting performance to the one of a 2AdditiveFactor Gaussian model as well as to the predictions of a dynamic NelsonSiegel model. 
Date:  2020–06 
URL:  http://d.repec.org/n?u=RePEc:arx:papers:2006.05750&r=all 
By:  Jan R. Magnus (Vrije Universiteit Amsterdam); Henk G.J. Pijls (University of Amsterdam); Enrique Sentana (CEMFI) 
Abstract:  We derive closedform expressions for the Jacobian of the matrix exponential function for both diagonalizable and defective matrices. The results are applied to two cases of interest in macroeconometrics: a continuoustime macro model and the parametrization of rotation matrices governing impulse response functions in structural vector autoregressions. 
Keywords:  Matrix differential calculus, Orthogonal matrix, Continuoustime Markov chain, OrnsteinUhlenbeck process 
JEL:  C65 C32 C63 
Date:  2020–06–20 
URL:  http://d.repec.org/n?u=RePEc:tin:wpaper:20200035&r=all 
By:  Constantin Bürgi; Dorine Boumans 
Abstract:  This paper introduces a new test of the predictive performance and market timing for categorical forecasts based on contingency tables when the user has noncategorical loss functions. For example, a user might be interested in the return of an underlying variable instead of just the direction. This new test statistic can also be used to determine whether directional forecasts are derived from nondirectional forecasts and whether point forecast have predictive value when transformed into directional forecasts. The tests are applied to the categorical exchange rate forecasts in the ifoInstitute’s World Economic Survey and to the point forecasts for quarterly GDP in the Philadelphia Fed's Survey of Professional Forecasters. We find that the loss function matters as exchange rate forecasters perform better under noncategorical loss functions, and the GDP forecasts have value up to two quarters ahead. 
Keywords:  contingency tables, categorical forecast, profitability, World Economic Survey, directional accuracy, market timing, forecast value 
JEL:  C12 C52 E37 F37 
Date:  2020 
URL:  http://d.repec.org/n?u=RePEc:ces:ceswps:_8266&r=all 
By:  Denis Fougère (CNRS  Centre National de la Recherche Scientifique, OSC  Observatoire sociologique du changement  Sciences Po  Sciences Po  CNRS  Centre National de la Recherche Scientifique, LIEPP  Laboratoire interdisciplinaire d'évaluation des politiques publiques [Sciences Po]  Sciences Po  Sciences Po, CEPR  Center for Economic Policy Research  CEPR, IZA  Forschungsinstitut zur Zukunft der Arbeit  Institute of Labor Economics); Nicolas Jacquemet (CES  Centre d'économie de la Sorbonne  UP1  Université PanthéonSorbonne  CNRS  Centre National de la Recherche Scientifique, PSE  Paris School of Economics) 
Abstract:  This paper describes, in a nontechnical way, the main impact evaluation methods, both experimental and quasiexperimental, and the statistical model underlying them. In the first part, we provide a brief survey of the papers making use of those methods that have been published by the journal Economie et Statistique / Economics and Statistics over the past fifteen years. In the second part, some of the most important methodological advances to have recently been put forward in this field of research are presented. To finish, we focus not only on the need to pay particular attention to the accuracy of the estimated effects, but also on the requirement to replicate evaluations, carried out by experimentation or quasiexperimentation, in order to distinguish false positives from proven effects. 
Keywords:  Causal effects,Causal inference,Evaluation methods 
Date:  2019–12 
URL:  http://d.repec.org/n?u=RePEc:hal:pseptp:hal02866828&r=all 