New Economics Papers
on Market Microstructure
Issue of 2008‒09‒20
five papers chosen by
Thanos Verousis

  1. Price Adjustment to News with Uncertain Precision By Nikolaus Hautsch; Dieter Hess; Christoph Müller
  2. Dynamic Factor Models for Multivariate Count Data: An Application to Stock-Market Trading Activity By Jung, Robert; Liesenfeld, Roman; Richard, Jean-Francois
  3. Volatility forecasting: the jumps do matter By Fulvio Corsi; Davide Pirino; Roberto Renò
  4. Curse of Mediocrity - On the Value of Asymmetric Fundamental Information in Asset Markets By Michael Kirchler
  5. Realisations of Finite-Sample Frequency-Selective Filters By D.S.G. Pollock

  1. By: Nikolaus Hautsch (Humboldt-Universität zu Berlin); Dieter Hess (University of Cologne); Christoph Müller (University of Cologne)
    Abstract: Bayesian learning provides the core concept of processing noisy information. In standard Bayesian frameworks, assessing the price impact of information requires perfect knowledge of news’ precision. In practice, however, precision is rarely disclosed. Therefore, we extend standard Bayesian learning, suggesting traders infer news’ precision from magnitudes of surprises and from external sources. We show that interactions of the different precision signals may result in highly nonlinear price responses. Empirical tests based on intra-day T-bond futures price reactions to employment releases confirm the model’s predictions and show that the effects are statistically and economically significant.
    Keywords: prediction Bayesian learning; macroeconomic announcements; information quality; precision signals
    JEL: E44 G14
    Date: 2008–06
  2. By: Jung, Robert; Liesenfeld, Roman; Richard, Jean-Francois
    Abstract: We propose a dynamic factor model for the analysis of multivariate time series count data. Our model allows for idiosyncratic as well as common serially correlated latent factors in order to account for potentially complex dynamic interdependence between series of counts. The model is estimated under alternative count distributions (Poisson and negative binomial). Maximum Likelihood estimation requires high-dimensional numerical integration in order to marginalize the joint distribution with respect to the unobserved dynamic factors. We rely upon the Monte-Carlo integration procedure known as Efficient Importance Sampling which produces fast and numerically accurate estimates of the likelihood function. The model is applied to time series data consisting of numbers of trades in 5 minutes intervals for five NYSE stocks from two industrial sectors. The estimated model accounts for all key dynamic and distributional features of the data. We find strong evidence of a common factor which we interpret as reflecting market-wide news. In contrast, sector-specific factors are found to be statistically insignifficant.
    Keywords: Dynamic latent variables, Importance sampling, Mixture of distribution models, Poisson distribution, Simulated Maximum Likelihood
    Date: 2008
  3. By: Fulvio Corsi; Davide Pirino; Roberto Renò
    Abstract: This study reconsiders the role of jumps for volatility forecasting by showing that jumps have positive and mostly significant impact on future volatility. This result becomes apparent once volatility is correctly separated into its continuous and discontinuous component. To this purpose, we introduce the concept of threshold multipower variation (TMPV), which is based on the joint use of bipower variation and threshold estimation. With respect to alternative methods, our TMPV estimator provides less biased and robust estimates of the continuous quadratic variation and jumps. This technique also provides a new test for jump detection which has substantially more power than traditional tests. We use this separation to forecast volatility by employing an heterogeneous autoregressive (HAR) model which is suitable to parsimoniously model long memory in realized volatility time series. Empirical analysis shows that the proposed techniques improve significantly the accuracy of volatility forecasts for the S&P500 index, single stocks and US bond yields, especially in periods following the occurrence of a jump
    Keywords: volatility forecasting, jumps, bipower variation, threshold estimation, stock, bond
    JEL: G1 C1 C22 C53
    Date: 2008–06
  4. By: Michael Kirchler
    Abstract: In this paper we present results from experimental asset markets and simulations with traders who receive asymmetric information about the fundamental value of an asset. In the experimental markets with repetition insiders outperform the market and uninformed random traders perform equally well as average informed traders. This is in line with the results of the equilibrium simulation output in which traders choose be- tween a random strategy and their fundamental strategy. We further ¯nd that the persistent underperformance of the average informed is not due to their overcon¯dence but due to the asymmetric information structure of the market.
    Keywords: Information economics, experimental economics, agent-based model, overconfidence, value of information
    JEL: C91 C92 G14
    Date: 2008–09
  5. By: D.S.G. Pollock
    Abstract: A filtered data sequence can be obtained by multiplying the Fourier ordinates of the data by the ordinates of the frequency response of the filter and by applying the inverse Fourier transform to carry the product back to the time domain. Using this technique, it is possible, within the constraints of a finite sample, to design an ideal frequency-selective filter that will preserve all elements within a specified range of frequencies and that will remove all elements outside it. Approximations to ideal filters that are implemented in the time domain are commonly based on truncated versions of the infinite sequences of coefficients derived from the Fourier transforms of rectangular frequency response functions. An alternative to truncating an infinite sequence of coefficients is to wrap it around a circle of a circumference equal in length to the data sequence and to add the overlying coefficients. The coefficients of the wrapped filter can also be obtained by applying a discrete Fourier transform to a set of ordinates sampled from the frequency response function. Applying the coefficients to the data via circular convolution produces results that are identical to those obtained by a multiplication in the frequency domain, which constitutes a more efficient approach.
    Keywords: Signal extraction; Linear filtering; Frequency-domain analysis
    Date: 2008–09

This issue is ©2008 by Thanos Verousis. 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.