{"repo":"Jebel-Quant/basanos","free":true,"listed":false,"github":"https://github.com/Jebel-Quant/basanos","clone":"git clone https://github.com/Jebel-Quant/basanos.git","description":"Implementing a first hurdle for expected returns","language":"Python","stars":18,"topics":["feature-analysis","portfolio-optimization","quantitative-finance","signals"],"license":"MIT","category":"trading","readme_excerpt":"Basanos Correlation-aware portfolio optimization and analytics for Python. --- Basanos computes correlation-adjusted risk positions from price data and expected-return signals. It estimates time-varying EWMA correlations, applies shrinkage towards the identity matrix, and solves a normalized linear system per timestamp to produce stable, scale-invariant positions — implementing a first hurdle for expected returns. Table of Contents - Idea - Features - Installation - Quick Start - Notebooks - Documentation - Development - License Idea Most systematic strategies produce a raw signal vector μ — one number per asset indicating how bullish or bearish the model is. Sizing each position in direct proportion to its signal ignores the fact that correlated assets will receive large, overlapping bets in the same direction, concentrating risk rather than diversifying it. Basanos treats position sizing as a linear system : where C is the (shrunk, time-varying) correlation matrix and μ is the signal. Solving for x inverts the correlation structure — assets that share a lot of co-movement with the rest of the portfolio receive smaller positions, while idiosyncratic assets can carry more. The result is a set of risk positions that express the full information in the signal while respecting the portfolio's correlation geometry. Three design choices keep the output stable and usable in practice: 1. EWMA estimates — both volatility and correlations are computed as exponentially weighted moving ","default_branch":null,"files":null,"tree":[],"storefront":"/r/Jebel-Quant","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/Jebel-Quant/basanos/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}