Quantitative Investment Analysis by Type-2 Fuzzy Random Support Vector Regression

Abstract

Financial markets are connected well these days. One class assets’ price performance is usually affected by movements of other classes of assets. The liquidity conduction mechanism usually is: if capital surface of money market is tight, investors may dump short-term treasury bonds to exchange additional liquidities. It may affects the performance of treasury bonds’ yield of maturity. Credit bonds’ return rate thus will level up hysteretic. Financing cost of companies accordingly raise, then throws effects on their stock prices. However, situation changes along with increase in complexity of markets’ behaviors these days. In order to model movements of assets’ price performance, analysis of linkage between different markets is thus becoming more and more important. Nothing like stock market, money market or bond market is an over-the-counter market, where assets’ prices are often presented in the form of classes of discrete quotations with trader’s subjective judgements, thus are hard to analyze. Given concern to this, we define the Type 2 fuzzy random variable (T2 fuzzy random variable) to quantify those bid/offer behaviours in this paper. Moreover, we build a T2 fuzzy random support vector regression scheme to study relationships between these markets. T2 fuzzy random support vector regression is developed from traditional support vector regression and is able to cope with fuzzy data, which has less computation complexity and better generalization performance than linear algorithms.