G17 - Financial Forecasting and SimulationReturn

One of the main objectives of this study is to develop a qualitative model that will serve the decision makers' CFOs (chief financial officers), where, as a rule, it is decided without deeper processing of information many factors that affect each other significantly. Lack of appropriate statistical information in connection with turbulently changing environments suggests that further research is needed to extend existing IPO models based on statistical analyzes. The paper is using basic qualitative research of trends. A qualitative trend model can be developed under conditions when the relevant quantitative model must be heavily simplified. The key information input into IPO is expert knowledge.  The solution of a trend model M(X) is a set S of scenarios where X is the set of n variables quantified by the trends. All possible transitions among the scenarios S are generated. An oriented transitional graph G has as nodes the set of scenarios S and as arcs the transitions T. An oriented G path describes any possible future and past time behaviour of the IPO system under study. The case study presents a model based on integration of equationless relations using 8 variables e.g. Market condition, Recognisability or Liquidity risk.  There are 17 scenarios S and 41 transitions T among them. All pairs of relationships are based on trends, either increasing, constant, or decreasing. The key input of the correct IPO timing analysis is based on the knowledge of experts traced from qualitative heuristics. The transition graph is a qualitative interpretation of all possible quantitative time series of all variables used in our IPO timetable and should be used as an effective tool to support CFO decisions.

The article is focused on the portfolio returns modelling, to be specific on the estimation of VaR as a portfolio risk measure. The goal of the article is to discover, how the characteristics of a market evolution influence the accuracy of VaR estimation of the internationally diversified portfolio of stock indices. Returns of portfolio are modelled by NIG model coupled together by Student copula function. The VaR estimation is back tested in different periods. The volatility of returns in each considered period evolves differently. The accuracy of VaR estimation is measured on the basis of the number of observed exceptions and its comparison to expected number of exceptions. It is apparent from the results, that for constant volatility period the model is accurate. When volatility changes in considered period, the model underestimates or overestimates the VaR.