Currently, various applications for Neural Networks and Machine Learning have been identified. In particular, their application to investment models seeks to identify patterns in historical financial data and use them to predict future trends, thus making investment decisions.
Numa Americas Corp is one of the few family office in the US and the only one in Peru that offers investment strategies based on Machine Learning.
About our Machine Learning Lachesis strategy
- The model is composed of two neural network models with each containing approximately 158 inputs. These inputs are based on various technical indicators such as Price, Volume, Volatility, and a combination of them.
- The Strategy decides how the allocation should be invested during the week. It has 3 possible outputs: Short, Neutral & Long. Short positions allows Lachesis to earn gains when market drops.
- ML Lachesis offers Portfolio Hedging. The model chooses to hold less market exposure at times of low certainty.
- Over the time, Lachesis has consistently overperformed the market.
Machine Learning Lachesis Return vs Benchmarks
Benchmark = 25% SPY + 25% QQQ + 25% IWM + 25% MDY.
All information for ML prior to July 2019 is back-tested, based on the methodology that was in effect on the Inception Date.
