nep-ecm New Economics Papers
on Econometrics
Issue of 2017‒08‒20
twelve papers chosen by
Sune Karlsson
Örebro universitet

  1. Is Industrial Production Still the Dominant Factor for the US Economy? By Andreou, Elena; Gagliardini, Patrick; Ghysels, Eric; Rubin, Mirco
  2. A GED-based regression to fit the actual data distribution By Mattera, Raffaele
  3. Multi-scale analysis of lead-lag relationships in high-frequency financial markets By Takaki Hayashi; Yuta Koike
  4. A Stochastic Recurrence Equation Approach to Stationarity and phi-Mixing of a Class of Nonlinear ARCH Models By Francisco (F.) Blasques; Marc Nientker
  5. A General Class of Multifractional Processes and its Application to Cross-listing Stocks By Qidi Peng; Ran Zhao
  6. On the Role of Covariates in the Synthetic Control Method By Botosaru, Irene; Ferman, Bruno
  7. Indirect Inference with Importance Sampling: An Application to Women’s Wage Growth By Robert M. Sauer; Christopher R. Taber
  8. Optimal Estimation of Multi-Country Gaussian Dynamic Term Structure Models Using Linear Regressions By Antonio Diez de los Rios
  9. On the overestimation of the largest eigenvalue of a covariance matrix By Soufiane Hayou
  10. Mixed Causal-Noncausal Autoregressions with Strictly Exogenous Regressors By Hecq, Alain; Issler, João Victor; Telg, Sean
  11. Opportunities and Challenges: Lessons from Analyzing Terabytes of Scanner Data By Serena Ng
  12. The Returns to Personality Traits across the Wage Distribution By Matthias Collischon

  1. By: Andreou, Elena; Gagliardini, Patrick; Ghysels, Eric; Rubin, Mirco
    Abstract: We propose a new class of large approximate factor models which enable us to study the full spectrum of quarterly Industrial Production (IP) sector data combined with annual non-IP sectors of the economy. We derive the large sample properties of the estimators and test statistics for the new class of unobservable factor models involving mixed frequency data and common as well as frequency-specific factors. Despite the growth of service sectors, we find that a single common factor explaining 90% of the variability in IP output growth index also explains 60% of total GDP output growth fluctuations. A single low frequency factor unrelated to manufacturing explains 14% of GDP growth. The picture with a structural factor model featuring technological innovations is quite different. Last but not least, our identification and inference methodology rely on novel results on group factor models that are of general interest beyond the mixed frequency framework and the application of the paper.
    Keywords: GDP growth; Group Factor models; MIDAS
    JEL: C32 C33 C38 E32
    Date: 2017–08
  2. By: Mattera, Raffaele
    Abstract: The classical literature of econometrics emphasizes the useful role of normal distribution behind the parameters’ estimate in linear regression model and the OLS method is one of the most important tools used by economist and econometricians to study the magnitude and the relationship between economic phenomena. However, the OLS method has too strict assumptions and the estimates resulting from its use could be improved by developing new statistical analysis methods. In particular, the assumption of the Gaussian distribution of errors can be eliminated, and it is possible to develop regression analysis using different estimators. Using the GED function, it is possible to estimate the parameters with the GED estimators, which are unbiased. The aim of this paper is to demonstrate that GED method is theoretically a better solution than OLS method for the regression analysis because the main propriety of this copula is the perfectly adaptive to actual distributions of data. Not forcing the distribution of dataset, we can obtain better estimates. To demonstrated it, we have presented a simulation study. Another important result of this paper is the building of a new generalized goodness-of-fit indicator, called r-squared of order p.
    Keywords: Generalized Error Distribution, kurtosis indexes, maximum likelihood, regression analysis, goodness of fit, r-squared of order p
    JEL: C1 C13 C18
    Date: 2017–08–14
  3. By: Takaki Hayashi; Yuta Koike
    Abstract: We propose a novel estimation procedure for scale-by-scale lead-lag relationships of financial assets observed at a high-frequency in a non-synchronous manner. The proposed estimation procedure does not require any interpolation processing of the original data and is applicable to quite fine resolution data. The validity of the proposed estimators is shown under the continuous-time framework developed in our previous work Hayashi and Koike (2016). An empirical application shows promising results of the proposed approach.
    Date: 2017–08
  4. By: Francisco (F.) Blasques (VU Amsterdam, The Netherlands; Tinbergen Institute, The Netherlands); Marc Nientker (VU Amsterdam, The Netherlands)
    Abstract: This article generalises the results of Sadi and Zakoian (2006) to a considerably larger class of nonlinear ARCH models with discontinuities, leverage effects and robust news impact curves. We propose a new method of proof for the existence of a strictly stationary and phi-mixing solution. Moreover, we show that any path converges to this solution. The proof relies on stochastic recurrence equation theory and builds on the work of Bougerol (1993) and Straumann (2005). The assumptions that we need for this approach are less restrictive than those imposed in Sadi and Zakoian (2006) and typically found in Markov chain theory, as they require very little from the distribution of the underlying process. Furthermore, they can be stated in a general setting for random functions on a separable Banach space as is done in Straumann and Mikosch (2006). Finally, we state sufficient conditions for the existence of moments.
    Keywords: Ergodicity; GARCH-type models; mixing; nonlinear time series; stationarity,stochastic recurrence equations; threshold models
    JEL: C50 C51 C58
    Date: 2017–08–02
  5. By: Qidi Peng; Ran Zhao
    Abstract: We introduce a general class of multifractional stochastic processes driven by a multifractional Brownian motion and study estimation of their pointwise H\"older exponents (PHE) using the localized generalized quadratic variation approach. By comparing it with the other two benchmark estimation approaches through a simulation study, we show that our estimator has better performance in the case where the observed process is some unknown bivariate function of time and multifractional Brownian motion. The time-varying PHE feature allows us to apply such class of multifractional processes to model stock prices under various market conditions. An empirical study on modeling cross-listed stocks provides new evidence that equity's path roughness varies via time and price informativeness properties from global markets.
    Date: 2017–08
  6. By: Botosaru, Irene; Ferman, Bruno
    Abstract: This note revisits the role of time-invariant observed covariates in the Synthetic Control (SC) method. We first derive conditions under which the original result of Abadie et al (2010) regarding the bias of the SC estimator remains valid when we relax the assumption of a perfect match on observed covariates and assume only a perfect match on pre-treatment outcomes. We then show that, even when the conditions for the first result are valid, a perfect match on pre-treatment outcomes does not generally imply an approximate match for all covariates. This will only be true for those that are both relevant and whose effects (over time) are not collinear with the effects of other observed and unobserved covariates. Taken together, our results show that a perfect match on covariates should not be required for the SC method, as long as there is a perfect match on a long set of pre-treatment outcomes.
    Keywords: Synthetic controls, covariates, perfect match
    JEL: C13 C21 C23
    Date: 2017–08–14
  7. By: Robert M. Sauer; Christopher R. Taber
    Abstract: This paper has two main parts. In the first, we describe a method that smooths the objective function in a general class of indirect inference models. Our smoothing procedure makes use of importance sampling weights in estimation of the auxiliary model on simulated data. The importance sampling weights are constructed from likelihood contributions implied by the structural model. Since this approach does not require transformations of endogenous variables in the structural model, we avoid the potential approximation errors that may arise in other smoothing approaches for indirect inference. We show that our alternative smoothing method yields consistent estimates. The second part of the paper applies the method to estimating the effect of women’s fertility on their human capital accumulation. We find that the curvature in the wage profile is determined primarily by curvature in the human capital accumulation function as a function of previous human capital, as opposed to being driven primarily by age. We also find a modest effect of fertility induced nonemployment spells on human capital accumulation. We estimate that the difference in wages among prime age women would be approximately 3% higher if the relationship between fertility and working were eliminated.
    JEL: C51 J16
    Date: 2017–08
  8. By: Antonio Diez de los Rios
    Abstract: This paper proposes a novel asymptotic least-squares estimator of multi-country Gaussian dynamic term structure models that is easy to compute and asymptotically efficient, even when the number of countries is relatively large—a situation in which other recently proposed approaches lose their tractability. We illustrate our estimator within the context of a seven-country, 10-factor term structure model.
    Keywords: Asset pricing; Econometric and statistical methods; Exchange rates; Interest rates
    JEL: E43 F31 G12 G15
    Date: 2017
  9. By: Soufiane Hayou
    Abstract: In this paper, we use a new approach to prove that the largest eigenvalue of the sample covariance matrix of a normally distributed vector is bigger than the true largest eigenvalue with probability 1 when the dimension is infinite. We prove a similar result for the smallest eigenvalue.
    Date: 2017–08
  10. By: Hecq, Alain; Issler, João Victor; Telg, Sean
    Abstract: The mixed autoregressive causal-noncausal model (MAR) has been proposed to estimate economic relationships involving explosive roots in their autoregressive part, as they have stationary forward solutions. In previous work, possible exogenous variables in economic relationships are substituted into the error term to ensure the univariate MAR structure of the variable of interest. To allow for the impact of exogenous fundamental variables directly, we instead consider a MARX representation which allows for the inclusion of strictly exogenous regressors. We develop the asymptotic distribution of the MARX parameters. We assume a Student's t-likelihood to derive closed form solutions of the corresponding standard errors. By means of Monte Carlo simulations, we evaluate the accuracy of MARX model selection based on information criteria. We investigate the influence of the U.S. exchange rate and the U.S. industrial production index on several commodity prices.
    Keywords: Mixed causal-noncausal process, non-Gaussian errors, identification, rational expectation models, commodity prices
    JEL: C22 E31 E37
    Date: 2017–08–11
  11. By: Serena Ng
    Abstract: This paper seeks to better understand what makes big data analysis different, what we can and cannot do with existing econometric tools, and what issues need to be dealt with in order to work with the data efficiently. As a case study, I set out to extract any business cycle information that might exist in four terabytes of weekly scanner data. The main challenge is to handle the volume, variety, and characteristics of the data within the constraints of our computing environment. Scalable and efficient algorithms are available to ease the computation burden, but they often have unknown statistical properties and are not designed for the purpose of efficient estimation or optimal inference. As well, economic data have unique characteristics that generic algorithms may not accommodate. There is a need for computationally efficient econometric methods as big data is likely here to stay.
    JEL: C81
    Date: 2017–08
  12. By: Matthias Collischon
    Abstract: This paper investigates heterogeneous wage effects of non-cognitive skills across the wage distribution. I develop a model of wage determination under uncertainty with respect to individual productivity based on three components (minimum wages, productivity premiums, bargaining premiums). Based on this model, I expect (i) a larger importance and (ii) larger effects of non-cognitive skills for high-wage employees compared to their low-wage counterparts. I test these hypotheses with unconditional quantile regressions using large-scale survey data from Germany, the UK, and Australia. To test the joint explanatory contribution of multiple variables within a quantile-regression framework, I propose a new statistic that quantifies the rise in explanatory power generated by additional explanatory variables. The findings indicate a rising importance as well as increasing effects of certain personality traits (agreeableness, neuroticism and risk taking) across the wage distribution for full-time employed males and females.
    Keywords: non-cognitive skills, personality traits, unconditional quantile regression
    JEL: C21 J24 J31
    Date: 2017

This nep-ecm issue is ©2017 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.