nep-ets New Economics Papers
on Econometric Time Series
Issue of 2007‒04‒28
six papers chosen by
Yong Yin
SUNY at Buffalo

  1. Online Forecast Combination for Dependent Heterogeneous Data By Sancetta, A.
  2. Markov switching GARCH models of currency turmoil in Southeast Asia By Celso Brunetti; Roberto S. Mariano; Chiara Scotti; Augustine H.H. Tan
  3. Bayesian Estimation of the Markov-Switching GARCH(1,1) Model with Student-t Innovations By David Ardia
  4. Dynamic Modeling of Large Dimensional Covariance Matrices By Valeri Voev
  5. Panel Intensity Models with Latent Factors: An Application to the Trading Dynamics on the Foreign Exchange Market¤ By Ingmar Nolte; Valeri Voev
  6. A Generalized ARFIMA Process with Markov-Switching Fractional Differencing Parameter By Wen-Jen Tsay; Wolfgang Härdle

  1. By: Sancetta, A.
    Abstract: This paper studies a procedure to combine individual forecasts that achieve theoretical optimal performance. The results apply to a wide variety of loss functions and no conditions are imposed on the data sequences and the individual forecasts apart from a tail condition. The theoretical results show that the bounds are also valid in the case of time varying combination weights, under specific conditions on the nature of time variation. Some experimental evidence to confirm the results is provided.
    Keywords: Forecast Combination, Model Selection, Multiplicative Update, Non-asymptotic Bound, On-line Learning.
    JEL: C53 C14
    Date: 2007–04
  2. By: Celso Brunetti; Roberto S. Mariano; Chiara Scotti; Augustine H.H. Tan
    Abstract: This paper analyzes exchange rate turmoil with a Markov Switching GARCH model. We distinguish between two different regimes in both the conditional mean and the conditional variance: "ordinary" regime, characterized by low exchange rate changes and low volatility, and "turbulent" regime, characterized by high exchange rate movements and high volatility. We also allow the transition probabilities to vary over time as functions of economic and financial indicators. We find that real effective exchange rates, money supply relative to reserves, stock index returns, and bank stock index returns and volatility contain valuable information for identifying turbulence and ordinary periods.
    Date: 2007
  3. By: David Ardia (Department of Quantitative Economics)
    Abstract: This article proposes the Bayesian estimation of the MSGARCH model with Student-t innovations. We introduce a new MCMC scheme which generates the GARCH parameters by block, the vector of state variables in a multi-move manner and the degrees of freedom parameter of the Student-t distribution using an efficient rejection technique. Our methodology is fully automatic and avoids the time-consuming and difficult task of choosing and tuning a sampling algorithm. As an application, we fit a single-regime GARCH model and a MSGARCH model to SMI log-returns. We use the random permutation sampler to find suitable identification constraints for the MSGARCH model and show the presence of two distinct volatility regimes in the time series. By using the Deviance information criterion and estimating the model likelihoods we show the in-sample superiority of the MSGARCH model. Finally, we test the forecasting performance of the competing models based on the VaR and document the superiority of the Markov-switching specification.
    Keywords: Bayesian;MCMC;Markov-switching;HMM;GARCH;DIC;Marginal likelihood;Model likelihood;VaR;SMI
    JEL: C11 C13 C15 C22 C52 C53
    Date: 2007–04–12
  4. By: Valeri Voev (University of Konstanz)
    Abstract: Modelling and forecasting the covariance of financial return series has always been a challange due to the so-called "curse of dimensionality". This paper proposes a methodology that is applicable in large dimensional cases and is based on a time series of realized covariance matrices. Some solutions are also presented to the problem of non-positive definite forecasts. This methodology is then compared to some traditional models on the basis of its forecasting performance employing Diebold-Mariano tests. We show that our approach is better suited to capture the dynamic features of volatilities and covolatilities compared to the sample covariance based models.
    Date: 2007–02–01
  5. By: Ingmar Nolte (University of Konstanz); Valeri Voev (University of Konstanz)
    Abstract: We develop a panel intensity model, with a time varying latent factor, which captures the influence of unobserved time effects and allows for correlation across individuals. The model is designed to analyze individual trading behavior on the basis of trading activity datasets, which are characterized by four dimensions: an irregularly-spaced time scale, trading activity types, trading instruments and investors. Our approach extends the stochastic conditional intensity model of Bauwens & Hautsch (2006) to panel duration data. We show how to estimate the model parameters by a simulated maximum likelihood technique adopting the efficient importance sampling approach of Richard & Zhang (2005). We provide an application to a trading activity dataset from an internet trading platform in the foreign exchange market and we find support for the presence of behavioral biases and discuss implications for portfolio theory.
    Keywords: Trading Activity Datasets, Panel Intensity Models, Latent Factors, Efficient Importance Sampling, Behavioral Finance
    JEL: G10 F31 C32
    Date: 2007–02–28
  6. By: Wen-Jen Tsay; Wolfgang Härdle
    Abstract: We propose a general class of Markov-switching-ARFIMA processes in order to combine strands of long memory and Markov-switching literature. Although the coverage of this class of models is broad, we show that these models can be easily estimated with the DLV algorithm proposed. This algorithm combines the Durbin-Levinson and Viterbi procedures. A Monte Carlo experiment reveals that the finite sample performance of the proposed algorithm for a simple mixture model of Markov-switching mean and ARFIMA(1, d, 1) process is satisfactory. We apply the Markov-switching-ARFIMA models to the U.S. real interest rates, the Nile river level, and the U.S. unemployment rates, respectively. The results are all highly consistent with the conjectures made or empirical results found in the literature. Particularly, we confirm the conjecture in Beran and Terrin (1996) that the observations 1 to about 100 of the Nile river data seem to be more independent than the subsequent observations, and the value of differencing parameter is lower for the first 100 observations than for the subsequent data.
    Keywords: Markov chain; ARFIMA process; Viterbi algorithm; Long memory.
    JEL: C14 C22 C32 C52 C53 G12
    Date: 2007–04

This nep-ets issue is ©2007 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.