Μέθοδοι υπολογισμού της αξίας - σε - κίνδυνο μεικτών χαρτοφυλακίων
The definition and implementation of Value at Risk (VAR) is one of the most modern and most popular risk assessment tools in the world of investment universe and risk management. The purpose of this thesis is the study of four alternative techniques of VAR estimation and the development of a simple tool for estimating VAR, in Excel format, which can be used by private investors. The techniques applied in this study are (1) the historical simulation method, (2) Modified VAR, (3) a method based on EWMA model and (4) a method by using Monte Carlo simulation. As part of the analysis, we have examined seven alternative assets for the period 2005-2014. In the beginning, we describe the concept and application of the VAR measure on the market risk assessment, with extensive analysis of the theoretical basis and literature review on the comparison of alternative techniques. The empirical study concerns the assessment of alternatives VAR methods and compare the results, through the construction of an integrated tool for the measurement of market risk. Emphasis is given to the reduction brought in the VAR diversification of a portfolio, which we call the benefit of diversification. The findings are summarized in an over-estimation of the VAR using the EWMA method, but the most important is that the historical method appears to offer the most underestimated results. Finally, the analysis of mixed portfolios shown that adding bonds in an equity portfolio, significantly reduces the value at risk.