On Tuesday June, 19 2018 a scientific seminar on analysis of financial time series with application of quantile regression and Bayesian technique took place.

The paper  ‘A Bayesian quantile time series model for asset returns’ was presented by dr Gelly Mitrodima from the Department of Statistics at London School of Economics and Political Science (UK). Many researchers from the Department of Econometrics and Statistics, Department of Applications of Informatics and Mathematics in Economics and many guests from other centers for econometrics in Poland took part in this event. The seminar was led by Professor Magdalena Osińska from our Faculty.

A Bayesian quantile time series model for asset returns - abstract:

The conditional distribution of asset returns has been widely  studied in the literature using a wide range of models for the conditional time-varying variance or volatility. However, empirical  studies show that other features of the distribution of asset returns may also vary over time. Our aim is to study the time variation in the return distribution beyond volatility, described by a collection of conditional quantiles. Direct modelling of quantile for Bayesian inference is challenging, since it involves analytic expressions for both the quantile function and its inverse to define the likelihood. Thus, we propose a novel class of Bayesian non-parametric priors for quantiles built around a random transformation. This allows fast and efficient Markov chain Monte Carlo (MCMC) methods to be applied for posterior simulation and forecasting. Under this framework, we avoid strong parametric assumptions about the underlying distribution, and so we obtain a model that is flexible about the shape of the distribution. We define a stationary model and we derive the stationary mean and variance of the quantiles. In our empirical exercise, we find that the model fits the data well, offers robust results, and acceptable forecasts for a sample of stock, index, and commodity returns.