nep-ecm New Economics Papers
on Econometrics
Issue of 2018‒04‒30
nine papers chosen by
Sune Karlsson
Örebro universitet

  1. On the Estimation of Treatment Effects with Endogenous Misreporting By Nguimkeu, Pierre; Denteh, Augustine; Tchernis, Rusty
  2. Using the Area Under an Estimated ROC Curve to Test the Adequacy of Binary Predictors By Robert Pal Lieli; Yu-Chin Hsu
  3. A new approach for detecting shifts in forecast accuracy By Chiu, Ching-Wai (Jeremy); hayes, simon; kapetanios, george; Theodoridis, Konstantinos
  4. An Internally Consistent Approach to the Estimation of Market Power and Cost Efficiency with an Application to U.S. Banking By Tsionas, Mike; Malikov, Emir; Kumbhakar, Subal C.
  5. R2 bounds for predictive models: what univariate properties tell us about multivariate predictability By Stephen Wright; James Mitchell; Donald Robertson
  6. Skating on Thin Evidence: Implications for Public Policy By Doucouliagos, Chris; Paldam, Martin; Stanley, T. D.
  7. Does Gentrification Displace Poor Households? An 'Identification-Via-Interaction' Approach By Sevrin Waights
  8. GEL-Based Inference from Unconditional Moment Inequality Restrictions By Nicky L. Grant; Richard J. Smith
  9. Economic predictions with big data: the illusion of sparsity By Giannone, Domenico; Lenza, Michele; Primiceri, Giorgio E.

  1. By: Nguimkeu, Pierre (Georgia State University); Denteh, Augustine (Georgia State University); Tchernis, Rusty (Georgia State University)
    Abstract: Participation in social programs is often misreported in survey data, complicating the estimation of the effects of those programs. In this paper, we propose a model to estimate treatment effects under endogenous participation and endogenous misreporting. We show that failure to account for endogenous misreporting can result in the estimate of the treatment effect having an opposite sign from the true effect. We present an expression for the asymptotic bias of both OLS and IV estimators and discuss the conditions under which sign reversal may occur. We provide a method for eliminating this bias when researchers have access to information related to both participation and misreporting. We establish the consistency and asymptotic normality of our estimator and assess its small sample performance through Monte Carlo simulations. An empirical example is given to illustrate the proposed method.
    Keywords: endogeneity, misclassification, treatment effect, binary regressor, partial observability, bias
    JEL: C35 C51
    Date: 2018–03
  2. By: Robert Pal Lieli; Yu-Chin Hsu
    Abstract: We consider using the area under an empirical receiver operating characteristic (ROC) curve to test the hypothesis that a predictive index combined with a range of cutoffs performs no better than pure chance in forecasting a binary outcome. This corresponds to the null hypothesis that the area in question, denoted as AUC, is 1/2. We show that if the predictive index comes from a first stage regression model estimated over the same data set, then testing the null based on standard asymptotic normality results leads to severe size distortion in general settings. We then analytically derive the proper asymptotic null distribution of the empirical AUC in a special case; namely, when the first stage regressors are Bernoulli random variables. This distribution can be utilized to construct a fully in-sample test of H0 : AUC = 1=2 with correct size and more power than out-of-sample tests based on sample splitting, though practical application becomes cumbersome with more than two regressors.
    Date: 2018–03–19
  3. By: Chiu, Ching-Wai (Jeremy) (Bank of England); hayes, simon (Bank of England); kapetanios, george (Kings College); Theodoridis, Konstantinos (Cardiff University)
    Abstract: Forecasts play a critical role at inflation-targeting central banks, such as the Bank of England. Breaks in the forecast performance of a model can potentially incur important policy costs. Commonly used statistical procedures, however, implicitly put a lot of weight on type I errors (or false positives), which result in a relatively low power of tests to identify forecast breakdowns in small samples. We develop a procedure which aims at capturing the policy cost of missing a break. We use data-based rules to find the test size that optimally trades off the costs associated with false positives with those that can result from a break going undetected for too long. In so doing, we also explicitly study forecast errors as a multivariate system. The covariance between forecast errors for different series, though often overlooked in the forecasting literature, not only enables us to consider testing in a multivariate setting but also increases the test power. As a result, we can tailor the choice of the critical values for each series not only to the in-sample properties of each series but also to how the series for forecast errors covary.
    Keywords: Forecast breaks; statistical decision making; central banking
    JEL: C53 E47 E58
    Date: 2018–04–13
  4. By: Tsionas, Mike; Malikov, Emir; Kumbhakar, Subal C.
    Abstract: We develop a novel unified econometric methodology for the formal examination of the market power -- cost efficiency nexus. Our approach can meaningfully accommodate a mutually dependent relationship between the firm's cost efficiency and marker power (as measured by the Lerner index) by explicitly modeling the simultaneous determination of the two in a system of nonlinear equations consisting of the firm's cost frontier and the revenue-to-cost ratio equation derived from its stochastic revenue function. Our framework places no a priori restrictions on the sign of the dependence between the firm's market power and efficiency as well as allows for different hierarchical orderings between the two, enabling us to discriminate between competing quiet life and efficient structure hypotheses. Among other benefits, our approach completely obviates the need for second-stage regressions of the cost efficiency estimates on the constructed market power measures which, while widely prevalent in the literature, suffer from multiple econometric problems as well as lack internal consistency/validity. We showcase our methodology by applying it to a panel of U.S. commercial banks in 1984-2007 using Bayesian MCMC methods.
    Keywords: Productivity, Competitiveness, Efficiency, Market Power, Lerner Index, Banks, Quiet Life Hypothesis
    JEL: C11 C30 D24 D40 G21
    Date: 2018
  5. By: Stephen Wright (Birkbeck, University of London); James Mitchell (Warwick Business School); Donald Robertson (University of Cambridge)
    Abstract: A longstanding puzzle in macroeconomic forecasting has been that a wide variety of multivariate models have struggled to out-predict univariate models consistently. We seek an explanation for this puzzle in terms of population properties. We derive bounds for the predictive R2 of the true, but unknown, multivariate model from univariate ARMA parameters alone. These bounds can be quite tight, implying little forecasting gain even if we knew the true multivariate model. We illustrate using CPI inflation data and the Eurozone in a specification motivated by a preferred-habitat model to test for monetary policy transmission domestically and internationally. Our findings suggest an impact of monetary policy on variance processes only and provides evidence for an international channel of monetary transmission on both money and capital markets. This is, to our knowledge, the first attempt to use search-engine data in the context of monetary policy.
    Keywords: attention, internet search, Google, monetary policy, ECB, FED, international financial markets, macro-finance, sovereign bonds, international finance, bond markets, preferred habitat models.
    JEL: C22 C32 C53 E37
    Date: 2018–04
  6. By: Doucouliagos, Chris (Deakin University); Paldam, Martin (Aarhus University); Stanley, T. D. (Deakin University)
    Abstract: Good public policy needs to be evidence based. However, the evidence base is thin for many policy issues. How can policy makers best respond to such thin areas of research that are also quite likely to change over time? Our survey investigates the evolution of the econometric evidence base for 101 economic issues, drawing upon 42,578 effect sizes (mainly elasticities and correlations) from 4,300 econometric studies. We evaluate the performance of six approaches to early research assessment: the simple unweighted mean; the median; the Paldam, "divide by 2" rule of thumb; the unrestricted weighted least squares (WLS) weighted average; the PET-PEESE meta-regression correction for publication bias; the weighted average of the adequately powered (WAAP); and WAAP-WLS. Lowest prediction errors are found in the Paldam rule of thumb and WLS. WLS typically reduces the initial exaggeration of thin evidence by half.
    Keywords: thin evidence, meta-regression, WLS, WAAP, Paldam rule of thumb
    JEL: C1 H00 H5
    Date: 2018–03
  7. By: Sevrin Waights
    Abstract: My theoretical model motivates an 'identification-via-interaction' (IvI) approach that separates the causal impact of gentrification on poor exits from endogenous channels. In the empirical analysis, I create a measure of gentrification as the increase in the share of neighbourhood residents who hold a university degree based on the UK Census for 1991, 2001 and 2011. Applying the IvI approach for a sample of private renters from the British Household Panel Survey, 1991-2008, I find that gentrification results in displacement. The IvI approach has general applications in estimating causal relationships where variables are highly endogenous.
    Keywords: neighbourhood change, mobility, turnover, causality, cities, urban, housing
    JEL: R21 R23 R31 C20
    Date: 2018–04
  8. By: Nicky L. Grant; Richard J. Smith
    Date: 2018
  9. By: Giannone, Domenico (Federal Reserve Bank of New York); Lenza, Michele (European Central Bank and ECARES); Primiceri, Giorgio E. (Northwestern University, CEPR, and NBER)
    Abstract: We compare sparse and dense representations of predictive models in macroeconomics, microeconomics, and finance. To deal with a large number of possible predictors, we specify a prior that allows for both variable selection and shrinkage. The posterior distribution does not typically concentrate on a single sparse or dense model, but on a wide set of models. A clearer pattern of sparsity can only emerge when models of very low dimension are strongly favored a priori.
    Keywords: model selection; shrinkage; high dimensional data
    JEL: C11 C53
    Date: 2018–04–01

This nep-ecm issue is ©2018 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.