Audrius Kabašinskas

The risk–return profile of Lithuanian private pension funds

Abstract The introduction of a private pension funds in conjunction with the public social security system is the essence of pension system reform that was implemented in Lithuania. The performance of private funds is mainly presented by fund’s net asset value and few classical risk estimates. Such evaluation shows the management company’s ability to profitably invest funds, but does not give the evidential risk–return evaluation. This paper refers to the overall statistical analysis of 26 private pension funds over a certain time period. The objective of the research is to determine the risk–return profile of pension funds and to answer the question whether the categories specified based on investment strategy in equities reflect fund’s empirical behaviour. Research methodology includes the statistical analysis, risk measuring, performance ratio estimation, and K-means clustering. The conclusions obtained by the research allow determining whether the distinct pension funds have beaten a low risk reference and are adequately assigned to a certain risk category.

The construction of an investment portfolio using stochastic programming

ABSTRACT The aim of this paper is to construct a portfolio of eight different stocks from New York Stock Exchange market (AIR, ABM, TSCO, HLX, KO, DIS, AMZN, and VZ) using stochastic programming. The next stage (period) prices are generated using a stochastic difference equation in order to introduce uncertainty. For the portfolio selection, we use three different risk measures – min–max decision rule, value-at-risk, and conditional value-at-risk. After constructing three different portfolios, they are compared using well-known efficiency ratios – Sharpe, Sortino, and Rachev ratios.

Accounting and governance risk forecasting in the health care industry

Previous authors have proved the advantage of commercial Accounting and Governance Risk (AGR) evaluation methods over academic methods. However, the information used in commercial methods is not readily available to an investor. Therefore, the most important features used in academic methods and the AGR was forecast by Random Forests. It found a weak relation between the AGR rating and share price data (Close and Volume), using a skew t-distribution. For visualisation we used the Kohonen map, which identified three clusters. Clusters revealed AGR increasing, decreasing trendsetting and cluster-based companies which appear to have no clear trend. A self-organised map (SOM) used the AGR history of alpha-stable distribution parameters, which were calculated from the stock data (Close and Volume). Also, the test sample (companies rating data), following from skew t-distribution, has been simulated by maximum likelihood method, and parameters of the skew t-distribution have been estimated.