A comparison of financial duration models via density forecasts

Bauwens, Luc; Giot, Pierre; Grammig, Joachim; Veredas, David

doi:10.1016/j.ijforecast.2003.09.014

Cited by 160 publications

(118 citation statements)

References 19 publications

Supporting

Mentioning

109

Contrasting

Unclassified

Order By: Relevance

“…Moreover, we do not assume that the intraday patterns are the same across different days of the week but do estimation for each of them separately. This is also different from most of previous studies, some of which allow but do not detect weakly seasonal components, e.g., Bauwens and Giot (2000), although Grammig and Wellner (2002) and Bauwens, Giot, Grammig, and Veredas (2004) also condition both on timeof-day and on day-of-week. An extreme position is taken in Meitz and Teräsvirta (2003), where splines are fit separately for each day of the sample.…”

Section: Seasonal Data Adjustmentcontrasting

confidence: 99%

“…models. We follow the algorithm of evaluating density forecasts proposed by Diebold, Gunther, and Tay (1998) and adapted by Bauwens, Giot, Grammig, and Veredas (2004) to duration data. In a nutshell, to test the specification of conditional distribution, one may generate a sequence of probability integral transforms…”

Section: Density Forecasting Resultsmentioning

confidence: 99%

“…Appropriate testing tools include Pearson's goodness-of-fit statistic for uniformity and Ljung-Box statistics for serial uncorrelatedness of the z-sequence and of its squares; in addition, visual inspection of histograms of z is able to provide informal evidence about possible causes of model failure (for more details, see Bauwens, Giot, Grammig, and Veredas, 2004).…”

Section: Density Forecasting Resultsmentioning

confidence: 99%

“…Using the class of so-called Skewed Generalized Error (SGE) distributions (Bali and Theodossiou, 2003) to model the conditional distribution within the Log-ARMA-GARCH framework allows one to adequately capture slight right skewness and slim-tailedness of log-durations of the MICEx stocks. In addition, density forecasting experiments (as in Bauwens, Giot, Grammig, and Veredas, 2004) show that the SGE Log-ARMA-GARCH model performs out-of-sample no worse than some preferable ACD specifications. We also analyze what factors determine the dynamics of log-durations and in which way.…”

mentioning

confidence: 96%

See 3 more Smart Citations

Trade intensity in the Russian stock market: dynamics, distribution and determinants

Anatolyev

Shakin

2007

Applied Financial Economics

View full text Add to dashboard Cite

We investigate the distribution and evolution of intertrade durations for frequently traded stocks at the Moscow Interbank Currency Exchange. We use a flexible econometric model based on ARMA and GARCH which, when coupled with a certain class of distributions that allow for skewness and slim-tailedness, adequately captures the characteristics of conditional distribution of durations for Russian stocks, and is able to generate high quality density forecasts. We also analyze what factors determine the dynamics of logdurations and in which way. The results in particular indicate that the Russian market is characterized by aggressive informed traders and timid liquidity traders, and that the participants react evenly to upward and downward short-run price trends.

show abstract

Section: Seasonal Data Adjustmentcontrasting

confidence: 99%

Section: Density Forecasting Resultsmentioning

confidence: 99%

Section: Density Forecasting Resultsmentioning

confidence: 99%

mentioning

confidence: 96%

See 2 more Smart Citations

Trade intensity in the Russian stock market: dynamics, distribution and determinants

Anatolyev

Shakin

2007

Applied Financial Economics

View full text Add to dashboard Cite

show abstract

“…Using the representation of the measurement density as an integral involving the -function in (11), it follows that the …ltered density of the state variable at time t = 1 may be expressed as…”

Section: 22mentioning

confidence: 99%

Non-parametric estimation of forecast distributions in non-Gaussian, non-linear state space models

Ng¹,

Forbes²,

Martin³

et al. 2013

International Journal of Forecasting

View full text Add to dashboard Cite

The object of this paper is to produce non-parametric maximum likelihood estimates of forecast distributions in a general non-Gaussian, non-linear state space setting. The transition densities that de…ne the evolution of the dynamic state process are represented in parametric form, but the conditional distribution of the non-Gaussian variable is estimated non-parametrically. The …ltering and prediction distributions are estimated via a computationally e¢ cient algorithm that exploits the functional relationship between the observed variable, the state variable and a measurement error with an invariant distribution. Simulation experiments are used to document the accuracy of the non-parametric method relative to both correctly and incorrectly speci…ed parametric alternatives. In an empirical illustration, the method is used to produce sequential estimates of the forecast distribution of realized volatility on the S&P500 stock index during the recent …nancial crisis. A resampling technique for measuring sampling variation in the estimated forecast distributions is also demonstrated.

show abstract

References

2010

ARCH Models for Financial Applications

View full text Add to dashboard Cite

A comparison of financial duration models via density forecasts

Cited by 160 publications

References 19 publications

Trade intensity in the Russian stock market: dynamics, distribution and determinants

Trade intensity in the Russian stock market: dynamics, distribution and determinants

Non-parametric estimation of forecast distributions in non-Gaussian, non-linear state space models

References

Contact Info

Product

Resources

About