Signed graphs for risk management

Abstract

Modern portfolio theory suggests that investor can reduce their investment risk by selecting a basket of investment assets that has collectively lower risk than any individual assets. Selection of investment assets is based on the expected rate of return and an expected risk measure for the portfolio and these are closely knitted to the degree of correlation between assets in the portfolio.

On a related matter, the notion of structural balance for signed graphs in the context of portfolio analysis involves the use of degree of correlation between the assets in the portfolio. Hence, our project first identifies balanced and unbalanced signed graphs in the context of investment portfolios. Trading strategy is then deduced based on the notion. Thereafter, back-testing on historical securities prices to investigate the profitability is carried out. The author has specifically design a program to perform the task of back-testing.