nep-ecm New Economics Papers
on Econometrics
Issue of 2015‒09‒11
ten papers chosen by
Sune Karlsson
Örebro universitet

  1. Dirty spatial econometrics By Giuseppe Arbia; Giuseppe Espa; Diego Giuliani
  2. Three essays in econometric theory By Gan, Zhuojiong
  3. Is a normal copula the right copula? By Amengual, Dante; Sentana, Enrique
  4. Quasifiltering for time-series modeling By Tsyplakov, Alexander
  5. Publication Bias in Meta-Analysis: Confidence Intervals for Rosenthal’s Fail-Safe Number By Fragkos, Konstantinos C.; Tsagris, Michail; Frangos, Christos C.
  6. Steady-state priors and Bayesian variable selection in VAR forecasting By Dimitrios P. Louzis
  7. Structural Analysis with Multivariate Autoregressive Index Models By Carreiro, Andrea; Kapetanios, George; Marcellino, Massimiliano
  8. Optimal Bandwidth Selection for the Fuzzy Regression Discontinuity Estimator By Yoichi Arai; Hidehiko Ichimura
  9. Experimenting with Measurement Error: Techniques with Applications to the Caltech Cohort Study By Ben Gillen; Erik Snowberg; Leeat Yariv
  10. Identification of Counterfactuals and Payoffs in Dynamic Discrete Choice with an Application to Land Use By Myrto Kalouptsidi; Paul T. Scott; Eduardo Souza-Rodrigues

  1. By: Giuseppe Arbia; Giuseppe Espa; Diego Giuliani
    Abstract: Spatial data are often contaminated with a series of imperfections that reduce their quality and can dramatically distort the inferential conclusions based on spatial econometric modeling. A ÒcleanÓ ideal situation considered in standard spatial econometrics textbooks is when we fit Cliff-Ord-type models to data where the spatial units constitute the full population, there are no missing data and there is no uncertainty on the spatial observations that are free from measurement and locational errors. Unfortunately in practical cases the reality is often very different and the datasets contain all sorts of imperfections: they are often based on a sample drawn from the whole population, some data are missing and they almost invariably contain both attribute and locational errors. This is a situation of ÒdirtyÓ spatial econometric modelling. Through a series of Monte Carlo experiments, this paper considers the effects on spatial econometric model estimation and hypothesis testing of two specific sources of dirt, namely missing data and locational errors.
    Date: 2015
  2. By: Gan, Zhuojiong (Tilburg University, School of Economics and Management)
    Abstract: This thesis consists of three essays in econometric theory. In the first essay, he considers a prediction problem with a large number of predictors. He improves the prediction precision of the standard factor model by allowing some variables to have idiosyncratic factors that are relevant for prediction. He selects idiosyncratic factors using a new model selection approach. In the second essay he studies two related tests of bivariate central symmetry. The asymptotic distributions of the two test statistics are established under rather weak conditions. He compares the finite sample performance of the test procedures with alternative tests by simulation. In the third and final essay, Zhuojiong proposes a pooled least-squares break point estimator for panel data. The estimator is shown to be consistent when the number of cross-sectional observations tends to infinity. Based on this break point estimator, he further proposes three consistent and asymptotically normally distributed slope estimators. The asymptotic variances of the three estimators are compared under a few simplifying assumptions.<br/>
    Date: 2015
  3. By: Amengual, Dante; Sentana, Enrique
    Abstract: We derive computationally simple and intuitive expressions for score tests of Gaussian copulas against Generalised Hyperbolic alternatives, including symmetric and asymmetric Student t, and Hermite polynomial expansions. We decompose our tests into third and fourth moment components, and obtain one-sided Likelihood Ratio analogues, whose asymptotic distribution we provide. We conduct Monte Carlo exercises to assess the finite sample properties of asymptotic and bootstrap versions of our tests. In an empirical application to CRSP stocks, we find that short-term reversals and momentum effects are better captured by non-Gaussian copulas. We estimate their parameters by indirect inference, and devise successful trading strategies.
    Keywords: Cokurtosis; Coskewness; indirect inference; Kuhn-Tucker test; momentum strategies; non-linear dependence; short-term reversals; Supremum test; underidentified parameters
    JEL: C12 C46 C52 G11 G14
    Date: 2015–09
  4. By: Tsyplakov, Alexander
    Abstract: In the paper a method for constructing new varieties of time-series models is proposed. The idea is to start from an unobserved components model in a state-space form and use it as an inspiration for development of another time-series model, in which time-varying underlying variables are directly observed. The goal is to replace a state-space model with an intractable likelihood function by another model, for which the likelihood function can be written in a closed form. If state transition equation of the parent state-space model is linear Gaussian, then the resulting model would belong to the class of score driven model (aka GAS, DCS).
    Keywords: time-series model, state-space model, score driven model
    JEL: C22 C51
    Date: 2015–07–10
  5. By: Fragkos, Konstantinos C.; Tsagris, Michail; Frangos, Christos C.
    Abstract: The purpose of the present paper is to assess the efficacy of confidence intervals for Rosenthal’s fail-safe number. Although Rosenthal’s estimator is highly used by researchers, its statistical properties are largely unexplored. First of all, we developed statistical theory which allowed us to produce confidence intervals for Rosenthal’s fail-safe number.This was produced by discerning whether the number of studies analysed in a meta-analysis is fixed or random. Each case produces different variance estimators. For a given number of studies and a given distribution, we provided five variance estimators. Confidence intervals are examined with a normal approximation and a nonparametric bootstrap. The accuracy of the different confidence interval estimates was then tested by methods of simulation under different distributional assumptions. The half normal distribution variance estimator has the best probability coverage. Finally, we provide a table of lower confidence intervals for Rosenthal’s estimator.
    Keywords: Meta-analysis, Rosenthal's fail safe number, file-drawer problem, bootstrap
    JEL: C18 C19
    Date: 2014–12
  6. By: Dimitrios P. Louzis (Bank of Greece)
    Abstract: This study proposes methods for estimating Bayesian vector autoregressions (VARs) with an automatic variable selection and an informative prior on the unconditional mean or steady-state of the system. We show that extant Gibbs sampling methods for Bayesian variable selection can be efficiently extended to incorporate prior beliefs on the steady-state of the economy. Empirical analysis, based on three major US macroeconomic time series, indicates that the out-of-sample forecasting accuracy of a VAR model is considerably improved when it combines both variable selection and steady-state prior information.
    Keywords: Bayesian VAR, Steady states, Variable selection, Macroeconomic forecasting
    JEL: C32
    Date: 2015–07
  7. By: Carreiro, Andrea; Kapetanios, George; Marcellino, Massimiliano
    Abstract: We address the issue of parameter dimensionality reduction in Vector Autoregressive models (VARs) for many variables by imposing specific reduced rank restrictions on the coefficient matrices that simplify the VARs into Multivariate Autoregressive Index (MAI) models. We derive the Wold representation implied by the MAIs and show that it is closely related to that associated with dynamic factor models. Next, we describe classical and Bayesian estimation of large MAIs, and discuss methods for the rank determination. Then, the theoretical analysis is extended to the case of general rank restrictions on the VAR coefficients. Finally, the performance of the MAIs is compared with that of large Bayesian VARs in the context of Monte Carlo simulations and two empirical applications, on on the transmission mechanism of monetary policy and the propagation of demand, supply and financial shocks.
    Keywords: Bayesian VARs; factor models; forecasting; large datasets; Multivariate Autoregressive Index models; Reduced Rank Regressions; structural analysis
    JEL: C11 C13 C33 C53
    Date: 2015–09
  8. By: Yoichi Arai (GRIPS); Hidehiko Ichimura (Department of Economics, University of Tokyo)
    Abstract: A new bandwidth selection method for the fuzzy regression discontinuity estimator is proposed. The method chooses two bandwidths simultaneously, one for each side of the cut-off point by using a criterion based on the estimated asymptotic mean square error taking into account a second-order bias term. A simulation study demonstrates the usefulness of the proposed method.
    Date: 2015–09
  9. By: Ben Gillen; Erik Snowberg; Leeat Yariv
    Abstract: Measurement error is ubiquitous in experimental work. It leads to imperfect statistical controls, attenuated estimated effects of elicited behaviors, and biased correlations between characteristics. We develop simple statistical techniques for dealing with experimental measurement error. These techniques are applied to data from the Caltech Cohort Study, which conducts repeated incentivized surveys of the Caltech student body. We illustrate the impact of measurement error by replicating three classic experiments, and showing that results change substantially when measurement error is taken into account. Collectively, these results show that failing to properly account for measurement error may cause a field-wide bias: it may lead scholars to identify "new" effects and phenomena that are actually similar to those previously documented.
    JEL: C81 C9 D8 J71
    Date: 2015–09
  10. By: Myrto Kalouptsidi; Paul T. Scott; Eduardo Souza-Rodrigues
    Abstract: Dynamic discrete choice models are non-parametrically not identified without restrictions on payoff functions, yet counterfactuals may be identified even when payoffs are not. We provide necessary and sufficient conditions for the identification of a wide range of counterfactuals for models with nonparametric payoffs, as well as for commonly used parametric functions, and we obtain both positive and negative results. We show that access to extra data of asset resale prices (when applicable) can solve non-identifiability of both payoffs and counterfactuals. The theoretical findings are illustrated empirically in the context of agricultural land use. First, we provide identification results for models with unobserved market-level state variables. Then, using a unique spatial dataset of land use choices and land resale prices, we estimate the model and investigate two policy counterfactuals: long run land use elasticity (identified) and a fertilizer tax (not identified, affected dramatically by restrictions).
    Keywords: Identification, Dynamic Discrete Choice, Counterfactual, Land Use
    JEL: C Q1
    Date: 2015–08–31

This nep-ecm issue is ©2015 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.