Filed: 10/09/2020
US Application No. 17/016,397

US Patent Issued: 08/09/2022

Tactical Investment Algorithms through Monte Carlo Backtesting

Methods and systems for computing optimized trading instructions, comprising applying a plurality of prediction models to compute trading instructions predicted to produce optimal outcomes based on received trading observations.

Filed: 10/09/2020
US Application No. 17/016,415

Applying Monte Carlo and Machine Learning Methods for Robust Convex Optimization-based Prediction Algorithms

Methods and systems for improving accuracy of convex optimization-based prediction models while reducing their computation load by clustering a plurality of received random variables.

Filed: 09/12/2019
US Application No. 62/899,164

Systems and Methods for a Factory that produces Tactical Investment Algorithms through Monte Carlo Backtesting

Layout of a strategy factory that produces tactical investment algorithms through a well-defined process.

Filed: 09/12/2019
US Application No. 62/899,163

Monte Carlo and Machine Learning Methods for Robust Convex Optimization

A platform that finds the most robust portfolio construction algorithm for a particular input structure.

Filed: 03/20/2019
US Application No. 16/358,937

Detecting False Positives in Statistical Models

A platform to derive the probability that a statistical discovery is a false positive, applying Monte Carlo and machine learning methods.

Filed: 05/27/2018
US Application No. 62/677,000

Framework for Reporting Investment Performance While Controlling for Selection Bias under Multiple Testing

A research surveillance platform that records all meta-data involved in the development of an investment strategy, so that the probability of a false discovery can be evaluated.

Filed: 03/29/2018
US Application No. 62/649,633

Detection of False Positives in Investment Strategies by Controlling for the Number of Trials Involved in a Discovery

False positives are detected by discounting the chances of encountering spurious patterns as a consequence of multiple testing and selection bias (DSR approach).

Filed: 03/22/2018
US Application No. 62/646,421

Detection of False Positives in Investment Strategies through Combinatorially-Purged Cross-Validation methods

False positives are detected by deriving a large distribution of alternative paths using combinatorial cross validation (see Chapter 12 here).

Filed: 09/29/2017
US Application No. 15/721,279

Hierarchical Construction of Investment Portfolios Using Clustered Machine Learning

HRP portfolios address three major concerns of quadratic optimizers in general and Markowitz's CLA in particular: Instability, concentration and opacity. Most notably, Monte Carlo experiments show that HRP portfolios deliver lower variance than CLA's out-of-sample, even though minimum-variance is CLA's objective function.

Filed: 05/09/2016
International Application No. PCT/IB2017/052703
US Application No. 62/333,484

Method and system for determining a weight allocation in a group comprising a plurality of items

A Quantum Computer is used to cluster a covariance matrix. The output is then used to build a robust diversified portolio.

Filed: 12/29/2015
US Application No. 15/391,764

Hierarchical Capital Allocation

We introduce a machine learning algorithm for the allocation of capital based on the hierarchical structure of the investment universe.

Filed: 03/28/2014
International Application No. PCT/US2015/023198, US Application No. 14/672,028 and US Application No. 15/904,523 

Systems and Methods for Crowdsourcing of Algorithmic Forecasting

We introduce new computational technologies that generate systematic investment portfolios based on forecasts contributed by a wide research collaboration network. At the same time, we control for the effects of selection bias under multiple testing.

Filed: 10/14/2011
US Application No. 13/273,958

Systems and methods for calculating an informed trading metric and applications thereof

Systems and computerized methods for calculating and using an informed trading metric are disclosed. The process of calculating the informed trading metric, executed by a processor, includes analyzing sets of trades of the security. The processor determines a magnitude of a difference between a volume of buy transactions and the volume of sell transactions in the plurality of trades. The processor then derives the informed trading metric based on the ratio of the determined difference magnitude to the total volume of analyzed trades. The derived informed trading metric may be employed by various systems to, among other things, hedge against market volatility, control securities exchange behavior, evaluate trader performance, and control the timing of trade execution by a broker dealer.

A system and related method for hedging risks associated with a market level of informed trading are disclosed. The system includes a network interface for receiving a data feed that includes informed trading metric data. The system also has a server for electronically publishing the informed trading metric data. A second network interface receives requests to purchase informed trading metric-based derivative contracts at a first price and offers to sell informed trading metric-based derivative contracts at a second price. A matching server matches the received requests with the received offers. A settlement processor settles the derivative contracts based on informed trading metric data received by the network interface.

A computerized exchange system and a method of operating a computerized exchange system are disclosed. The exchange system variably favors execution of buy trades or sell trades based on the value of an informed trading metric, thereby attempting to forestall predatory increases in order toxicity. The system includes a first network interface for receiving a plurality of securities trades, including a plurality of buy transactions and a plurality of sale transactions. A matching processor matches the securities trades to market makers. The matching processor is configured to obtain a value of the informed trading metric and to determine a buy transaction bias or a sell transaction bias based on the value of the informed trading metric. The matching processor then matches trades to market makers favoring buy transactions or sell transactions based on the determined bias. A settlement processor then settles the matched trades.