DYNAMIC VOLATILITY TARGETING

The recent financial crisis suggested us that investing in financial markets is a risky job and this turmoil has lead to a basic change in our understanding of financial markets and investment strategies. In particular the investment community has learned the lesson that markets are not always fully efficient and not rational as finance theory said; at the same time, investors have shifted from a mere return-based thinking to a risk-return based evaluation of investment opportunities. Diversification is neither an adequate nor trustworthy risk control mechanism: during crises asset classes that were uncorrelated in the past immediately they become correlated, for this reason market risk should be actively managed. Recently several investment banks and asset management companies with long-term investment horizons have responded with aversion to volatility by considering a number of risk control strategies. One approach of this kind is via derivative overlay strategies; however, our goal of stabilizing volatility can be achieved with risk-based asset allocation by taking advantage of two main market's features: the negative relationship between volatility and return and the persistence of volatility. In this thesis, rules-based investment strategies will be presented. They use risk indicators in the asset allocation process with two definite goals: in regular market conditions the strategy guarantees a 1:1 participation in the upside potential of financial markets making investors able to take advantage of any potential profits, in chaotic market conditions the investment strategy shifts assets away from risky assets into ¿safe havens¿ to protect investors from significant losses. The main idea of this asset allocation process is a strategy with two components: a risky asset (an equity index) that offers potential returns in the long run and a risk-free asset (short-term government bond index) where the investment amount can be shifted to whenever the equity market is in unfavorable conditions. The goal is to keep the portfolio volatility constant over time so during periods of volatility over the target we are going to decrease the exposure to the equity; in contrast, during periods of volatility below the target we are going to increase equity exposure and decrease the cash share. Because there are so many possible measures of an asset's volatility: historical standard deviation like Exponential Moving Average, Econometric models like GARCH models, Implied Volatility measures like VIX; the key to create a successful investment schemes is to find out an appropriate measure that can be used to trigger a re-balancing between equity and risk free. This thesis wants to compare different types of asset allocation based on different volatility measures to have a better idea of what is the risk measure that provides better performance and, at the same time, better protection for this type of risk-based strategy.