Search
× Search
Monday, December 23, 2024

Archived Discussions

Recent member discussions

The Algorithmic Traders' Association prides itself on providing a forum for the publication and dissemination of its members' white papers, research, reflections, works in progress, and other contributions. Please Note that archive searches and some of our members' publications are reserved for members only, so please log in or sign up to gain the most from our members' contributions.

Quantum Computing in Finance

photo

 Marcos Lopez de Prado, Senior Managing Director at GUGGENHEIM PARTNERS

 Wednesday, December 9, 2015

Multi-period portfolio optimization is a complex (indeed, NP-Complete) problem of great importance to asset managers. Quantum computers can solve this problem, with important savings in transaction costs, market impact and information leakage. * News: http://www.bloomberg.com/news/articles/2015-12-09/quantum-supercomputers-entice-wall-street-vowing-higher-returns * Presentation: http://ssrn.com/abstract=2694133 * Paper: http://ssrn.com/abstract=2649376


Print

10 comments on article "Quantum Computing in Finance"

photo

 Scott Boulette, Algorithmic Trading

 Thursday, December 10, 2015



@Marcos - I read the article but I wasn't clear on whether this is theory or actual practice at this point. Can you shed any light in this area?


photo

 Leonardo O., Researcher at Kadima Asset Management

 Friday, December 11, 2015



I fact, it's a grey area. Though there's already a quantum computer (DWave mentioned on the article), it's not yet complete acknowledged that it works the way it was supposed to. This month some guys from google published a paper stating that the machine is indeed a quantum computer (http://arxiv.org/abs/1512.02206). There are some discussion about the paper, but seems that soon we'll have some of these on our data centers


photo

 Konstantinos Skindilias, Senior Reseach Associate at ABM Analytics

 Sunday, December 13, 2015



I hope they put it at better use


photo

 Jenny Considine, Partner at Ossian Investments LP

 Sunday, December 13, 2015



The retirement security of our elderly citizens, after years of service to their children and countries.. What could be a better service


photo

 Jenny Considine, Partner at Ossian Investments LP

 Sunday, December 13, 2015



I saw a quantum computer at a trade show in NYC! Its fantastic, and very promising


photo

 Stephane Hardy, Computational Finance Quant and Options Trader

 Monday, December 14, 2015



If you need a quantum computer to earn a living, you need to look at you basic strategy and your bread and butter trades.

Bottle necks are usually trade quote delays and message routing delays.

But if you are continuously inverting data matrices or recompiling, to the point where your puter is too slow, why not split the tasks. Use a puter for each market segment, and serve results to other networked thread request.


photo

 Johann Christian L., CEO

 Monday, December 14, 2015



Portfolio optimization is not NP-complete - at least I do not a know NP-complete a portfolio optimization algorithm. Which algorithm do you mean?


photo

 Brian Wong, Assistant Portfolio Manager at Proprietary Trading Firm

 Wednesday, December 16, 2015



As we all know, financial time series is always full of uncertainties, the statistical properties of which evolves with time; the best adjust return figured out from computation is still restricted to be valid in the historical data only unless you correctly predict what these properties ( e.g volatility and correlation ) go on. Computation power is not a bottle-neck to me through cluster computing & algorithm complexity reduction; Perhaps, more powerful computers can tell me a better parameter and combination of portfolios in terms of historical data but it also increases the chance of curve-fitting if increasing unnecessary testing parameter space.


photo

 Konstantinos Skindilias, Senior Reseach Associate at ABM Analytics

 Thursday, December 17, 2015



@ Jane Considine: Good luck on that


photo

 Stephane Hardy, Computational Finance Quant and Options Trader

 Wednesday, January 20, 2016



Brian, viewing data arranged as a time series is but one way to view data. But you can also use first differences and higher moments to associate data, irrespective of time. For instance, align data irrespective of time, but according to acceleration, or relative acceleration with respect to non-singular information, like the Dow index. Then with a change of variable, you can identify causality, with time as a resulting solution that is elastic and allowed a flexible topology within your AI model. We are working with time and sales data, with volume, and within a published bid an ask envelope. There is a lot of info to be extracted here. There is more than moving averages and regressions.

The basic flaw here, is that these models always offer a direction.

Event driven AI does not give signals all the time. See "general covariance" also known as general relativity. Hope to hear from you.

Please login or register to post comments.

TRADING FUTURES AND OPTIONS INVOLVES SUBSTANTIAL RISK OF LOSS AND IS NOT SUITABLE FOR ALL INVESTORS
Terms Of UsePrivacy StatementCopyright 2018 Algorithmic Traders Association