nep-ets New Economics Papers
on Econometric Time Series
Issue of 2017‒04‒30
four papers chosen by
Yong Yin
SUNY at Buffalo

  1. Bias-corrected estimation of panel vector autoregressions By Geert Dhaene; Koen Jochmans
  2. Likelihood Inference in an Autoregression with Fixed Effects By Geert Dhaene; Koen Jochmans
  3. Do Bivariate Multifractal Models Improve Volatility Forecasting in Financial Time Series? An Application to Foreign Exchange and Stock Markets By Ruipeng Liu; Riza Demirer; Rangan Gupta; Mark E. Wohar
  4. Coherent Probabilistic Forecasts for Hierarchical Time Series By Souhaib Ben Taieb; James W. Taylor; Rob J. Hyndman

  1. By: Geert Dhaene (Université Catholique de Louvain); Koen Jochmans (Département d'économie)
    Abstract: We derive a bias-corrected least-squares estimator for panel vector autoregressions with fixed effects. The estimator is straightforward to implement and is asymptotically unbiased under asymptotics where the number of time series observations and the number of cross-sectional observations grow at the same rate. This makes the estimator particularly well suited for most macroeconomic data sets.
    Keywords: Bias Correction; Fixed Effects; Panel Data; Vector Autoregression
    JEL: C33
    Date: 2016–06
  2. By: Geert Dhaene (Université Catholique de Louvain); Koen Jochmans (Département d'économie)
    Abstract: We calculate the bias of the profile score for the regression coefficients in a multistratum autoregressive model with stratum-specific intercepts. The bias is free of incidental parameters. Centering the profile score delivers an unbiased estimating equation and, upon integration, an adjusted profile likelihood. A variety of other approaches to constructing modified profile likelihoods are shown to yield equivalent results. However, the global maximizer of the adjusted likelihood lies at infinity for any sample size, and the adjusted profile score has multiple zeros. Consistent parameter estimates are obtained as local maximizers inside or on an ellipsoid centered at the maximum likelihood estimator.
    Date: 2016–10
  3. By: Ruipeng Liu (Department of Finance, Deakin Business School, Deakin University, Melbourne, Australia); Riza Demirer (Department of Economics & Finance, Southern Illinois University Edwardsville, Edwardsville, USA); Rangan Gupta (Department of Economics, University of Pretoria, Pretoria, South Africa and IPAG Business School, Paris, France); Mark E. Wohar (College of Business Administration, University of Nebraska at Omaha, Omaha, USA and School of Business and Economics, Loughborough University, Leicestershire,UK)
    Abstract: This paper examines volatility linkages and forecasting for stock and foreign exchange (FX) markets from a novel perspective by utilizing a bivariate Markov-switching multifractal model (MSM) that accounts for possible interactions between stock and FX markets. Examining daily data from the advanced G6 and emerging BRICS nations, we compare the out-of-sample volatility forecasts from GARCH, univariate MSM and bivariate MSM models. Our findings show that the GARCH model generally offers superior volatility forecasts for short horizons, particularly for FX returns in advanced markets. Multifractal models, on the other hand, offer significant improvements for longer forecast horizons, consistently across most markets. Finally, the bivariate MF model provides superior forecasts compared to the univariate alternative in most G6 countries and more consistently for FX returns, while its benefits are limited in the case of emerging markets. Overall, our findings suggest that multifractal models can indeed improve out-of-sample volatility forecasts, particularly for longer horizons, while the bivariate specification can potentially extend the superior forecast performance to shorter horizons as well.
    Keywords: Long memory, multifractal models, simulation based inference, volatility forecasting, BRICS
    JEL: C11 C13 G15
    Date: 2017–04
  4. By: Souhaib Ben Taieb; James W. Taylor; Rob J. Hyndman
    Abstract: Many applications require forecasts for a hierarchy comprising a set of time series along with aggregates of subsets of these series. Although forecasts can be produced independently for each series in the hierarchy, typically this does not lead to coherent forecasts -- the property that forecasts add up appropriately across the hierarchy. State-of-the-art hierarchical forecasting methods usually reconcile the independently generated forecasts to satisfy the aggregation constraints. A fundamental limitation of prior research is that it has considered only the problem of forecasting the mean of each time series. We consider the situation where probabilistic forecasts are needed for each series in the hierarchy. We define forecast coherency in this setting, and propose an algorithm to compute predictive distributions for each series in the hierarchy. Our algorithm has the advantage of synthesizing information from different levels in the hierarchy through a sparse forecast combination and a probabilistic hierarchical aggregation. We evaluate the accuracy of our forecasting algorithm on both simulated data and large-scale electricity smart meter data. The results show consistent performance gains compared to state-of-the art methods.
    Keywords: forecast combination, probabilistic forecast, copula, machine learning.; Forecast combination, probabilistic forecast, copula, machine learning
    JEL: C53 Q47 C32
    Date: 2017

This nep-ets issue is ©2017 by Yong Yin. 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.