A new methodology that challenges the accepted shortcomings of other optimization frameworks
Traditional optimization methods assume that asset returns are normally distributed around a mean value; however, most assets, if not all, have “heavy-tailed” and skewed distributions that are not normal. Conventional methods based on normally distributed values have technical limitations that become apparent as the number of assets considered increases.
StableMetrics uses non-normal distributions within the sampling process to simulate and model more precise performance scenarios that in turn inform portfolio optimization decisions. In the resampling process, value-at-risk models account for simulated losses exceeding a defined threshold, enabling results designed to avoid excess risk exposure.
Fat-tailed and skewed asset forecasts can inform rather than complicate the allocation decision
StableMetrics upgrades the metric of asset variance in portfolio optimization. Enhance the value of proprietary estimates with precise asset modeling.
Our methodology estimates stable parameters and calculates density functions of an asset’s return distribution. The stable parameters permit customizable projected return distributions for each asset. The density function enables a multivariate return sampling process that recognizes the correlation of an asset’s returns in comparison both with its own unique distribution and with the distribution of each other asset in the portfolio. The approximations enable non-normal distribution analysis.
Pinpoint assets that drive performance, visualize the impact of market movements and report results with confidence
Register here to receive StableMetrics Feature Highlights.