“Buy low, sell high.” One cannot find a more succinct summary of a mean-reversion trading strategy; however, single assets that show stable mean-reversion over a significant period of time such that a mean-reversion trading strategy can readily become profitable are rare to find in markets today. Even if such gems were found, celebrating the discovery of the gateway to easy money could prove premature:
Posts
The Black-Litterman (BL) model is one of the many successfully used portfolio allocation models out there. Developed by Fischer Black and Robert Litterman at Goldman Sachs, it combines Capital Asset Pricing Theory (CAPM) with Bayesian statistics and Markowitz’s modern portfolio theory (Mean-Variance Optimisation) to produce efficient estimates of the portfolio weights.
Before getting into the nitty-gritty of the algorithm it is important to understand the motivations behind developing it and why is it favored by practitioners in the industry. For a long while, investors worked under the assumption that the risk and return relationship of a portfolio was linear, meaning that if an investor wanted higher returns, they would have to take on a higher level of risk.
This blog post explores the impacts of Covid-19 by simulating two investment portfolios – a portfolio consisting of peripheral stocks, versus a portfolio consisting of central stocks in the Planar Maximally Filtered Graph. The goal was to highlight the repercussions of the Covid related decline in the market, which shook the world in mid-February (in the case of the US markets). The portfolios take positions at the worst possible timing in order to understand – had you invested just before the dramatic crash of the market, how would a peripheral portfolio behave compared to a central portfolio? Are peripheral portfolios any better during an unexpected crisis?
Risk has always played a very large role in the world of finance with the performance of a large number of investment and trading strategies being dependent on the efficient estimation of underlying market risk. The covariance matrix is one of the most popular and widely used estimator of risk but due to its sensitivity to market conditions and dependence on historical data, it produces an unreliable estimation of true market risk. In this post, we go over some important methods of estimating covariance matrices which can be used in practice to remove noise from empirical estimates and produce better and reliable risk estimations.
For over half a century, most asset managers have used historical correlation matrices (empirical or factor-based) to develop investment strategies and build diversified portfolios. The Theory-Implied Correlation matrix combines external market views with emprirical values to generate new correlations which are less noisy and in sync with the economic theory.
As diversification is the only free lunch in finance, the Hierarchical Equal Risk Contribution Portfolio (HERC) aims at diversifying capital allocation and risk allocation. Briefly, the principle is to retain the correlations that really matter and once the assets are hierarchically clustered, a capital allocation is estimated. HERC allocates capital within and across the “right” number of clusters of assets at multiple hierarchical levels.
Throughout this post, we will explore the intuition behind Hierarchical Risk Parity and also learn to apply it using the MlFinLab library.
This article explores the intuition behind the development of Hierarchical Risk Parity, a detailed explanation of its working and how it compares to the other allocation algorithms.
The Hierarchical Risk Parity algorithm is fast, robust and flexible.