Power mixture forward performance processes

TL;DR

This paper develops a comprehensive framework for forward investment performance processes with power mixture initial conditions, including constant and stochastic risk aversion, and explores multi-investor portfolio management.

Contribution

It introduces a broad class of power mixture forward performance processes and characterizes two-power mixture processes with both constant and stochastic risk aversion.

Findings

Constructed a broad class of forward performance processes with power mixture initial conditions.

Fully characterized two-power mixture forward performance processes with constant risk aversion.

Analyzed the properties of stochastic risk aversion in two-power mixture processes.

Abstract

We consider the forward investment problem in market models where the stock prices are continuous semimartingales adapted to a Brownian filtration. We construct a broad class of forward performance processes with initial conditions of power mixture type, $u(x) = \int_{\mathbb{I}} \frac{x^{1-\gamma}}{1-\gamma }\nu(\mathrm{d} \gamma)$. We proceed to define and fully characterize two-power mixture forward performance processes with constant risk aversion coefficients in the interval $(0,1)$, and derive properties of two-power mixture forward performance processes when the risk aversion coefficients are continuous stochastic processes. Finally, we discuss the problem of managing an investment pool of two investors, whose respective preferences evolve as power forward performance processes.