nep-ets New Economics Papers
on Econometric Time Series
Issue of 2013‒05‒05
seven papers chosen by
Yong Yin
SUNY at Buffalo

  1. Multivariate high-frequency financial data via semi-Markov processes By Guglielmo D'Amico; Filippo Petroni
  2. Forecasting with Many Models: Model Confidence Sets and Forecast Combination By Jon D. Samuels; Rodrigo Sekkel
  3. Weak exogeneity in the financial point processes By Xu, Yongdeng
  4. The dynamics of trading duration, volume and price volatility – a vector MEM model By Xu, Yongdeng
  5. Comparing the Accuracy of Copula-Based Multivariate Density Forecasts in Selected Regions of Support By Cees Diks; Valentyn Panchenko; Oleg Sokolinskiy; Dick van Dijk
  6. Econometrics of co-jumps in high-frequency data with noise By Markus Bibinger; Lars Winkelmann; ;
  7. Non-parametric transformation regression with non-stationary data By Oliver Linton; Qiying Wang

  1. By: Guglielmo D'Amico; Filippo Petroni
    Abstract: In this paper we propose a bivariate generalization of a weighted indexed semi-Markov chains to study the high frequency price dynamics of traded stocks. We assume that financial returns are described by a weighted indexed semi-Markov chain model. We show, through Monte Carlo simulations, that the model is able to reproduce important stylized facts of financial time series like the persistence of volatility and at the same time it can reproduce the correlation between stocks. The model is applied to data from Italian stock market from 1 January 2007 until the end of December 2010.
    Date: 2013–05
  2. By: Jon D. Samuels; Rodrigo Sekkel
    Abstract: A longstanding finding in the forecasting literature is that averaging forecasts from different models often improves upon forecasts based on a single model, with equal weight averaging working particularly well. This paper analyzes the effects of trimming the set of models prior to averaging. We compare different trimming schemes and propose a new one based on Model Confidence Sets that take into account the statistical significance of historical out-of-sample forecasting performance. In an empirical application of forecasting U.S. macroeconomic indicators, we find significant gains in out-of-sample forecast accuracy from our proposed trimming method.
    Keywords: Econometric and statistical methods
    JEL: C53
    Date: 2013
  3. By: Xu, Yongdeng
    Abstract: This paper analyses issues related to weak exogeneity in a financial point process. We extend the Hausman test of weak exogeneity in a time series model and propose three cases in which weak exogeneity conditions will break down. The simulation study suggested that a failure of the exogeneity assumption implied biased estimators. The bias is very large in the third case non-weak exogeneity, which makes the econometric inferences on the parameters unreliable or even misleading. We then derive an LM test for weak exogeneity. The LM test is attractive because it only requires estimation of the restricted model. The empirical results indicate that the weak exogneity of duration is often rejected for frequently traded stocks, but is less likely to be rejected for infrequently traded stocks.
    Keywords: Weak exogeneity; ACD model; LM test; point process; market microstructure
    Date: 2013–04
  4. By: Xu, Yongdeng
    Abstract: We propose a general form of vector Multiplicative Error Model (MEM) for the dynamics of duration, volume and price volatility. The vector MEM relaxes the two restrictions often imposed by previous empirical work in market microstructure research, by allowing interdependence among the variables and relaxing weak exogeneity restrictions. We further propose a multivariate lognormal distribution for the vector MEM. The model is applied to the trade and quote data from the New York Stock Exchange (NYSE). The empirical results show that the vector MEM captures the dynamics of the trivariate system successfully. We find that times of greater activity or trades with larger size coincide with a higher number of informed traders present in the market. But we highlight that it is unexpected component of trading duration or trading volume that carry the information content. Moreover, our empirical results also suggest a significant feedback effect from price process to trading intensity, while the persistent quote changes and transient quote changes affect trading intensity in different direction, confirming Hasbrouck (1988,1991).
    Keywords: Vector MEM; ACD; GARCH; intraday trading process; duration; volume; volatility
    JEL: C15 C32 C52
    Date: 2013–04
  5. By: Cees Diks (CeNDEF, University of Amsterdam); Valentyn Panchenko (University of New South Wales); Oleg Sokolinskiy (Rutgers Business School); Dick van Dijk (Econometric Institute, Erasmus University Rotterdam)
    Abstract: This paper develops a testing framework for comparing the predictive accuracy of copula-based multivariate density forecasts, focusing on a specific part of the joint distribution. The test is framed in the context of the Kullback-Leibler Information Criterion, but using (out-of-sample) conditional likelihood and censored likelihood in order to focus the evaluation on the region of interest. Monte Carlo simulations document that the resulting test statistics have satisfactory size and power properties in small samples. In an empirical application to daily exchange rate returns we find evidence that the dependence structure varies with the sign and magnitude of returns, such that different parametric copula models achieve superior forecasting performance in different regions of the support. Our analysis highlights the importance of allowing for lower and upper tail dependence for accurate forecasting of common extreme appreciation and depreciation of different currencies.
    Keywords: Copula-based density forecast; Kullback-Leibler Information Criterion; out-of-sample forecast evaluation
    JEL: C12 C14 C32 C52 C53
    Date: 2013–04–19
  6. By: Markus Bibinger; Lars Winkelmann; ;
    Abstract: We establish estimation methods to determine co-jumps in multivariate high-frequency data with nonsynchronous observations and market microstructure noise. The ex-post quadratic covariation of the signal part, which is modeled by an Itˆo-semimartingale, is estimated with a locally adaptive spectral approach. Locally adaptive thresholding allows to disentangle the co-jump and continuous part in quadratic covariation. Our estimation procedure implicitly renders spot (co-)variance estimators. We derive a feasible stable limit theorem for a truncated spectral estimator of integrated covariance. A test for common jumps is obtained with a wild bootstrap strategy. We give an explicit guideline how to implement the method and test the algorithm in Monte Carlo simulations. An empirical application to intra-day tick-data demonstrates the practical value of the approach.
    Keywords: co-jumps, covolatility estimation, jump detection, microstructure noise, non-synchronous observations, quadratic covariation, spectral estimation, truncation
    JEL: C14 G32 E58
    Date: 2013–05
  7. By: Oliver Linton (Institute for Fiscal Studies and Cambridge University); Qiying Wang
    Abstract: We examine a kernel regression smoother for time series that takes account of the error correlation structure as proposed by Xiao et al. (2008). We show that this method continues to improve estimation in the case where the regressor is a unit root or near unit root process.
    Date: 2013–04

This nep-ets issue is ©2013 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.