We consider a mean-variance portfolio selection problem in financial market with contagion risk. The risky assets follow jump-diffusion model, which jumps are driven by multivariate Hawkes process mutual-excitation effect. feature of the captures risk sense that each price jump an asset increases likelihood future not only same but also other assets. apply stochastic maximum principle, backward...

Many problems in quantitative finance involve both predictive forecasting and decision-based optimization. Traditionally, models are optimized with unique prediction-based objectives constraints, therefore unaware of how those predictions will ultimately be used the context their final We present a stochastic optimization framework for integrating regression based mean-variance portfolio settin...

Probability-based distributions might be able to explain risk exposure well. Usually, one number, or at the very least, a limited number of numbers called key indicators (KRIs), are used describe level exposure. These values, which undeniably outcome specific model, frequently referred as essential critical indicators. Five indicators, including value-at-risk, tail variance, tail-value-at-risk,...

Reinforcement learning algorithms are typically geared towards optimizing the expected return of an agent. However, in many practical applications, low variance is desired to ensure reliability algorithm. In this paper, we propose on-policy and off-policy actor-critic that optimize a performance criterion involving both mean return. Previous work uses second moment estimate indirectly. Instead,...