Constrained Dynamic Mean-Variance Portfolio Selection in Continuous-Time

This paper revisits the dynamic MV portfolio selection problem with cone constraints in continuous-time. We first reformulate our constrained MV portfolio selection model into a special constrained LQ optimal control model and develop the optimal portfolio policy of our model. In addition, we provide an alternative method to resolve this dynamic MV portfolio selection problem with cone constraints. More specifically, instead of solving the correspondent HJB equation directly, we develop the optimal solution for this problem by using the special properties of value function induced from its model structure, such as the monotonicity and convexity of value function. Finally, we provide an example to illustrate how to use our solution in real application. The illustrative example demonstrates that our dynamic MV portfolio policy dominates the static MV portfolio policy.

Fundamental Analysis and Mean-Variance Optimal Portfolios

We integrate fundamental analysis with mean-variance portfolio optimization to form fully optimized fundamental portfolios. We find that fully optimized fundamental portfolios produce large out-of-sample factor alphas with high Sharpe ratios. They substantially outperform equal-weighted and value-weighted portfolios of stocks in the extreme decile of expected returns, an approach commonly used in fundamental analysis research. They also outperform the factor-based and parametric portfolio policy approaches used in the prior portfolio optimization literature. The relative performance gains from mean-variance optimized fundamental portfolios are persistent through time, robust to eliminating small capitalization firms from the investment set, and robust to incorporating estimated transactions costs. Our results suggest that future fundamental analysis research could implement this portfolio optimization approach to provide greater investment insights.

Hedging Demand in Long-Term Asset Allocation with an Application to Carry Trade Strategies*

We derive a closed-form expression for the mean and marginal hedging demand on risky assets in long-term asset allocation problems for individuals with constant relative risk aversion preferences. Our parametric portfolio policy rule accommodates an arbitrarily large number of state variables for predicting the state of nature and number of assets in the portfolio. The closed-form expression for the hedging demand is exact under polynomial specifications of the portfolio policy rule and a suitable approximation for unknown smooth parametric portfolio policy rules using Taylor expansions. The hedging demand on risky assets depends positively on the predictability of the risky asset and the persistence of the predictors, and negatively on the degree of investor’s relative risk aversion. We illustrate these insights empirically for a basket of currencies by showing the outperformance of rebalancing carry trade strategies over different investment horizons against a short-term (myopic) portfolio.

Relation-Aware Transformer for Portfolio Policy Learning

Portfolio selection is an important yet challenging task in AI for FinTech. One of the key issues is how to represent the non-stationary price series of assets in a portfolio, which is important for portfolio decisions. The existing methods, however, fall short of capturing: 1) the complicated sequential patterns for asset price series and 2) the price correlations among multiple assets. In this paper, under a deep reinforcement learning paradigm for portfolio selection, we propose a novel Relation-aware Transformer (RAT) to handle these aspects. Specifically, being equipped with our newly developed attention modules, RAT is structurally innovated to capture both sequential patterns and asset correlations for portfolio selection. Based on the extracted sequential features, RAT is able to make profitable portfolio decisions regarding each asset via a newly devised leverage operation. Extensive experiments on real-world crypto-currency and stock datasets verify the state-of-the-art performance of RAT.

A Portfolio Policy Package to Reduce Greenhouse Gas Emissions

Arguments for a portfolio of price, regulation and subsidy policy interventions to reduce the production and consumption of greenhouse gas emissions are presented. The operation and effects of each intervention are described and compared. A combination of different sets of market failures across the many potential decision changes available to producers and consumers to reduce emissions and different properties of the mitigation instruments support a portfolio approach to reduce emissions at a low cost.

Risk Taking and Fiscal Smoothing with Sovereign Wealth Funds in Advanced Economies

In an economy with a sovereign wealth fund (SWF), the government may draw on the fund to supplement other government revenues. If the fund is invested in risky assets, this introduces a new stochastic element into the government’s budget. We analyze the interaction between the draw from and risk taking in the SWF. Using non-expected utility preferences, we distinguish between intended changes and stochastic changes in the SWF draws over time. We show that the desire for smoothness in taxes and public services translates into smoothing of SWF draws and lower risk taking. It can even lead to procyclical rebalancing of the SWF portfolio. Future interest rates are associated with interest-rate risk. We show that this risk may lead to a higher optimal equity share in the SWF portfolio. Policy makers can use the draws from the SWF to smooth over time variation in risk-free rates.