Tea Poklepović

How to Measure Illiquidity on European Emerging Stock Markets

Abstract Background: Liquidity is, in practice of portfolio investment, an important attribute of stocks and measuring illiquidity presents a real challenge for researchers, primarily on developed stock markets. Moreover, there is a lack of research dealing with (il)liquidity on emerging markets. In the paper, the problem of applicability and validity of two well-known illiquidity measures, ILLIQ and TURN, on European emerging markets is observed. Objectives: The paper has two main purposes. The first is to test the relative performance of the two selected illiquidity measures in terms of their validity on European emerging stock markets. The second is to propose a new and improved illiquidity measure named Relative Change in Volume (RCV). Methods/Approach: Using daily returns and traded volumes for 12 stocks which are constituents of stock indices on seven observed markets, ILLIQ and TURN along with the new proposed measure are calculated and tested based on correlation with return. All measures are tested and proposed using the single stock approach. Results: It is shown that ILLIQ and TURN are not appropriate for seven observed markets. The measures do not follow the obligatory request that returns increase in illiquidity while RCV has the ability of taking into account the pressure of big differences in volume on return. RCV gives satisfactory results, making clear the distinction between liquid and illiquid stocks and between liquid and illiquid markets. Conclusions: The proposed measure potentially has important implications in illiquidity measurement in general, and not only for investors on emerging stock markets.

Neural Network Approach in Forecasting Realized Variance Using High-Frequency Data

Abstract Background: Since high-frequency data have become available, an unbiased volatility estimator, i.e. realized variance (RV) can be computed. Commonly used models for RV forecasting suffer from strong persistence with a high sensitivity to the returns distribution assumption and they use only daily returns. Objectives: The main objective is measurement and forecasting of RV. Two approaches are compared: Heterogeneous AutoRegressive model (HAR-RV) and Feedforward Neural Networks (FNNs). Even though HAR-RV-type models describe RV stylized facts very well, they ignore its nonlinear behaviour. Therefore, FNN-HAR-type models are developed. Methods/Approach: Firstly, an optimal sampling frequency with application to the DAX index is chosen. Secondly, in and out of sample predictions within HAR models and FNNs are compared using RMSE, AIC, the Wald test and the DM test. Weights of FNN-HAR-type models are estimated using the BP algorithm. Results: The optimal sampling frequency of RV is 10 minutes. Within HAR-type models, HAR-RV-J has better, but not significant, forecasting performances, while FNN-HAR-J and FNNLHAR- J have significantly better predictive accuracy in comparison to the FNN-HAR model. Conclusions: Compared to the traditional ones, FNN-HAR-type models are better in capturing nonlinear behaviour of RV. FNN-HAR-type models have better accuracy compared to traditional HAR-type models, but only on the sample data, whereas their out-of-sample predictive accuracy is approximately equal.