We continue to research quantum topics: portfolio optimization & quantum walks on graphs
July 11, 2021 - Top 20 CQNS most inefficient, 'down' picks & results
Medium Post (click to read)
May 25, 2021 - Top 10 CQNS 'up' picks & results
(We created > 13% alpha in 24 trading days)
Medium post (click to read)
Quick update (April 20, 2021) new
We have made some recent strides on our quantum and classical solvers.
1. On April 20, we optimized a stock portfolio of 142 stocks on the D-Wave Advantage 1.1 (vs. 132 before). It found valid and useful stock portfolios.
- In May improved our usage (and understanding) of 3 DWave classical samplers (gradient descent, simulated annealer, and TABU multi-start) and as of May 14, get useful answers from them.
- We worked with DWave to use a different codeset with the quantum annealer, and seem to get valid and useful solutions. Still testing this solution restructuring. Good news.
2. We added bespoke reverse simulated annealing to our set of classical proprietary solvers. We used to only go from hot to cold, and now we go from cold to hot to find you the deepest and best solutions to your QUBO or BQM problems.
3. We added a new Monte Carlo solver. We had random and fat tailed, and now we have a seeded Monte Carlo. This works better, finds deeper answers, and runs a little bit faster.
As always, our Particle Swarm Optimizer and our Simulated Bifurcator find independent solutions to our stock picking problems, and sometimes find the best solutions found. Most often, they find very good solutions, and add independence and diversity into the solutions we provide to our seeded Genetic Algorithm.
Our new server (from Dell) works great, and has significantly more processing power, storage and memory and runs at a much higher speed (all of the main components are larger, more powerful and faster). This helps us run our quantum algorithm deeper.
1. On April 20, we optimized a stock portfolio of 142 stocks on the D-Wave Advantage 1.1 (vs. 132 before). It found valid and useful stock portfolios.
- In May improved our usage (and understanding) of 3 DWave classical samplers (gradient descent, simulated annealer, and TABU multi-start) and as of May 14, get useful answers from them.
- We worked with DWave to use a different codeset with the quantum annealer, and seem to get valid and useful solutions. Still testing this solution restructuring. Good news.
2. We added bespoke reverse simulated annealing to our set of classical proprietary solvers. We used to only go from hot to cold, and now we go from cold to hot to find you the deepest and best solutions to your QUBO or BQM problems.
3. We added a new Monte Carlo solver. We had random and fat tailed, and now we have a seeded Monte Carlo. This works better, finds deeper answers, and runs a little bit faster.
As always, our Particle Swarm Optimizer and our Simulated Bifurcator find independent solutions to our stock picking problems, and sometimes find the best solutions found. Most often, they find very good solutions, and add independence and diversity into the solutions we provide to our seeded Genetic Algorithm.
Our new server (from Dell) works great, and has significantly more processing power, storage and memory and runs at a much higher speed (all of the main components are larger, more powerful and faster). This helps us run our quantum algorithm deeper.
Our latest paper (3 of 3): Portfolio Optimization of 3,171 Stocks
Authors: Jeffrey Cohen & Clark Alexander
Date: October 30, 2020
Date: October 30, 2020
Abstract: We analyze 3,171 US common stocks to create an efficient portfolio based on the Chicago Quantum Net Score (CQNS) and portfolio optimization. We begin with classical solvers and incorporate quantum annealing. We add a simulated bifurcator as a new classical solver and the new D-Wave Advantage(TM) quantum annealing computer as our new quantum solver.
15 pages, 9 figures.
Quantum Physics (quant-ph); Portfolio Management (q-fin.PM)
Cite as: arXiv:2011.01308 [quant-ph]
15 pages, 9 figures.
Quantum Physics (quant-ph); Portfolio Management (q-fin.PM)
Cite as: arXiv:2011.01308 [quant-ph]
Article 3 walk through & Nov 14, 2020 research update
Recent presentations of our work
|
Washington DC Quantum Computing Meetup: Sept 9 {completed}
"The race is on...Quantum Annealing vs. other methods we learned". A successful, 2-hour discussion. Zen4Quantum Meetup: Sept 18 {completed} "Quantum Stock Portfolios - Chicago Quantum Talk Series Part 2". An enjoyable 2-hour discussion where we discuss our goals, accomplishments, challenges, and newest methods. Recording here FS Club (London): Sept 21 {completed] "The Right Balance? Using a Quantum Annealing Computer for your Portfolio". A fast-paced 45 minute discussion on the the performance of published portfolios, the 'rig' we use, and our thoughts on scaling up further. Recording here Qubits 2020: Sept 28 {Completed} https://www.qubits.com/ Recording here IQT Europe: Oct 28 (Financial Services Panel Discussion) https://europe.iqtevent.com/ Recording here 
|
|
| ||
Our 2nd paper (2 of 3): Portfolio Optimization of 60 Stocks
Authors: Jeffrey Cohen, Alex Khan, Clark Alexander
Date: August 19, 2020
Abstract: We continue to investigate the use of quantum computers for building an optimal portfolio out of a universe of 60 U.S. listed, liquid equities. Starting from historical market data, we apply our unique problem formulation on the D-Wave Systems Inc. D-Wave 2000Q (TM) quantum annealing system (hereafter called D-Wave) to find the optimal risk vs return portfolio. We approach this first classically, then using the D-Wave, to select efficient buy and hold portfolios. Our results show that practitioners can use either classical or quantum annealing methods to select attractive portfolios. This builds upon our prior work on optimization of 40 stocks.
Comments:19 pages, 15 figures, 21 references
Subjects:General Finance (q-fin.GN); Quantum Physics (quant-ph)
Cite as:arXiv:2008.08669 [q-fin.GN]
Date: August 19, 2020
Abstract: We continue to investigate the use of quantum computers for building an optimal portfolio out of a universe of 60 U.S. listed, liquid equities. Starting from historical market data, we apply our unique problem formulation on the D-Wave Systems Inc. D-Wave 2000Q (TM) quantum annealing system (hereafter called D-Wave) to find the optimal risk vs return portfolio. We approach this first classically, then using the D-Wave, to select efficient buy and hold portfolios. Our results show that practitioners can use either classical or quantum annealing methods to select attractive portfolios. This builds upon our prior work on optimization of 40 stocks.
Comments:19 pages, 15 figures, 21 references
Subjects:General Finance (q-fin.GN); Quantum Physics (quant-ph)
Cite as:arXiv:2008.08669 [q-fin.GN]
Our 1st paper (1 of 3): Portfolio Optimization of 40 Stocks
Authors: Jeffrey Cohen, Alex Khan, Clark Alexander
Date: July 6, 2020
Abstract: We investigate the use of quantum computers for building a portfolio out of a universe of U.S. listed, liquid equities that contains an optimal set of stocks. Starting from historical market data, we look at various problem formulations on the D-Wave Systems Inc. D-Wave 2000Q(TM) System (hereafter called D-Wave) to find the optimal risk vs return portfolio; an optimized portfolio based on the Markowitz formulation and the Sharpe ratio, a simplified Chicago Quantum Ratio (CQR), then a new Chicago Quantum Net Score (CQNS). We approach this first classically, then by our new method on D-Wave. Our results show that practitioners can use a D-Wave to select attractive portfolios out of 40 U.S. liquid equities.
Comments:15 pages, 8 figures
Subjects:General Finance (q-fin.GN); Quantum Physics (quant-ph)
Cite as:arXiv:2007.01430 [q-fin.GN]
Date: July 6, 2020
Abstract: We investigate the use of quantum computers for building a portfolio out of a universe of U.S. listed, liquid equities that contains an optimal set of stocks. Starting from historical market data, we look at various problem formulations on the D-Wave Systems Inc. D-Wave 2000Q(TM) System (hereafter called D-Wave) to find the optimal risk vs return portfolio; an optimized portfolio based on the Markowitz formulation and the Sharpe ratio, a simplified Chicago Quantum Ratio (CQR), then a new Chicago Quantum Net Score (CQNS). We approach this first classically, then by our new method on D-Wave. Our results show that practitioners can use a D-Wave to select attractive portfolios out of 40 U.S. liquid equities.
Comments:15 pages, 8 figures
Subjects:General Finance (q-fin.GN); Quantum Physics (quant-ph)
Cite as:arXiv:2007.01430 [q-fin.GN]
