Nonparametric Methods in Asset Allocation

Equity Style Allocation: A Nonparametric Approach: The purpose of this paper is to produce a framework to assist with style allocation in Asian equity funds. We implement a nonparametric methodology to capture short-term stable time-varying relationships of otherwise long-term unstable relationships between numerous macroeconomic variables and style returns. We demonstrate that a nonparametric forecasting methodology produces positive performance after allowing for transaction costs, while the equivalent parametric forecasts are negative. The model can be implemented through tilting a funds style exposure to enhance performance. Even in the context of a long-only fund, the style exposures proposed by the model can be implemented as long-short exposures relative to a benchmark. Because the model is presented as a self financing market-neutral model, its implementation can be leveraged directly in a market-neutral fund or indirectly as (leveraged) style exposures in a long only fund.   We believe this paper to be the first paper to use a nonparametric methodology to assist in style switching. Other methodologies proposed in the literature have been used to assist in style switching in United States and European equity markets but to our knowledge none have been applied in the Asian equity markets as we do in this paper.

Hedge Fund Allocation: Evaluating Parametric and Nonparametric Forecasts Using Alternative Portfolio Construction Techniques: The objective of this paper is to propose a model to assist in constructing Asian funds of hedge funds. Using the ordinary least squares (OLS) regression, a nonparametric regression, and a nonlinear nonparametric (the simplex projection) approach to forecast hedge fund returns, we compare the accuracy of these forecasts. We perform a back-test to assess these forecasts using three different portfolio construction processes: an "optimized" portfolio, an equally-weighted portfolio, and the Kelly criterion-based portfolio. We find that the Kelly criterion is a reasonable method of constructing a fund of hedge funds and produces better results than a basic optimization or an equally-weighted portfolio construction method. We also find that the nonparametric forecasts and the OLS forecasts produce similar performance in a back-test at the hedge fund index level. At the individual fund level, our analysis indicates that the OLS forecasts produce higher directional accuracy than the nonparametric methods but the nonparametric methods produce more accurate forecasts than the OLS. In back-tests, the highest Sharpe ratio to predict hedge fund returns is achieved  through a combination of the OLS regression with the Fung-Hsieh eight-factor variables as predictors using the  Kelly criterion portfolio construction method. We also find that the combination of the nonparametric regression using the Fung-Hsieh eight-factor model variables as predictors of with the Kelly criterion portfolio construction method produces the highest Sharpe ratio. We find that in simulations using risk-adjusted total returns, the nonparametric regression generates superior Sharpe ratios than the analogous back-test using the OLS. However, the benefits of diversification plateau with portfolios of more than 20 hedge funds. These results generally hold with portfolio implementation lags up to 12 months. We believe this paper to be the first to evaluate nonparametric methodologies to assist in hedge fund allocation at the individual fund level for Asian hedge funds.