Optimal Feature Selection in Credit Scoring and Classification Using a Quantum Annealer

By Andrew Milne, Maxwell Rounds, & Phil Goddard

In credit scoring and classification, feature selection is used to reduce the number of variables input to a classifier. This can be done with a quadratic unconstrained binary optimization (QUBO) model, which attempts to select features that are both independent and influential. Quadratic optimization scales exponentially with the number of features, but a QUBO implementation on a quantum annealer has the potential to be faster than classical solvers. Tests were done using the German Credit Data from UC Irvine, and the results compared with those reported in the literature. In comparison with recursive feature elimination (RFE), a technique found in many software packages, QUBO Feature Selection yielded a smaller feature subset with no loss of accuracy. This opens up the possibility of using quantum annealers to programatically reduce the size of very large feature sets, especially as the size and availability of these devices increases.

An expanded version of this white paper has been published in: High-Performance Computing in Finance: Problems, Methods, and Solutions (2018)

WHITE PAPER

Most Recent Papers

CME Market Sentiment Meter Historical Market Analyses – Gold – 2019 Federal Funds Rate Cuts

Periods of Anxious market states for COMEX Gold futures (GC1) tended to be either short-lived or long-lived in the eight-year period ending in December 2019. In 2018, the U.S. saw economic growth and the Federal Reserve hiked rates four times during the year. The year was dominated primarily by Balanced market states, and GC consistently fell once the rate hikes were announced. The year 2019 saw slowed economic growth and increased tensions with China, Iran, and Russia. This led to a large rise in GC from July to November, and GC remained at a high relative to the beginning of the year. In 2019, the Federal Reserve made three rate cuts. During this time, the Market Sentiment Meter (MSM) indicated an extended period of Anxious market states from July to November.

Market Reactions to COVID-19

A review of Q1 2020 as seen in the CME Market Sentiment Meter By Anish R. Verma, Andrew Milne The COVID-19 pandemic had a notable effect on the eight futures and options products tracked by 1QBit’s CME Market Sentiment Meter. In some markets, such as U.S. equity index...

Quantum-Inspired Hierarchical Risk Parity

By Elham Alipour, Clemens Adolphs, Arman Zaribafiyan, & Maxwell Rounds
Posted on November 14, 2016

We present a quantum-inspired approach to portfolio optimization that is based on an optimization problem that can be solved using a quantum annealer. The proposed algorithm utilizes a hierarchical clustering tree that is based on the covariance matrix of the asset returns. We use real market data to benchmark our approach against other common portfolio optimization methods and demonstrate its strong performance in terms of a variety of risk measures and lower susceptibility to inaccuracies in the input data.