What does the team at Quantopian think of the use of Algebraic Topology to build globally stable trading algorithms? Are there any mathematicians on the Quantopian team who have studied this topic in some graduate math program and understand how to apply it to trading? Haven't seen much posting on this topic and the learning content seems to be focused on traditional statistical/linear approaches to data structure (even PCA assumes linearity in mapping from higher dimensions to lower ones, I believe). My intuition tells me that Ren Tech uses this math (in combination with number theory, attractors, etc) but the million dollar question is of course how do you combine all of it when building the algorithm. I have a few ideas for what could be done with Algebraic Topology in quant finance but wanted to hear a gut check from your team. Thank you for the response.
@Andy Y . Thank you for posting. Doing some checking around here, the Q team hasn't had any direct experience with Algebraic Topology and especially its application to trading. So, don't have any input one way or another. That said, we very much encourage you, and everyone in the community, to explore new approaches to finding alpha. Maybe you are on to something.
Please post something here in the forums if you do pursue this. We'd be excited to hear the results.
The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.
Hi Dan, I have continued on my search to use Algebraic Topology to find alpha. What I am trying to do is create persistence diagrams of time series data sets which might be indicative of a trading signal. In math terms, I would like to search for persistent homology invariants within the data. If I have the invariants I can then feed them into ML techniques to make predictions. I have spent some time understanding the theory and actual code that would be needed to write this into a trading algorithm, but the problem I have now with implementing it is that the only library (to my knowledge) which computes persistent homology is not supported on Quantopian Research Whitelist! The library is here https://pypi.org/project/scikit-tda/0.0.3/ (I need the Ripser.py 0.3.2 library).
I noticed that there is a way to work locally on my machine where I wouldnt have a restriction on which modules I can import, but then I don't have access your guys' trading data which is also not good. What work arounds can you suggest for me to be able to work with financial data and use this library? Can you guys whitelist that module? Thank you for your time!
Andy
@Dan @Quantopian Can anyone from Quantopian team please provide a response to my question? Trying to figure out how to move this idea forward if I can at all
@Andy ,
I've done some work in computational homology applied to logic minimization.
I haven't done anything with the new concepts of persistent homology, yet they look interesting.
I've been working on using zipline away from Q's platform, both for live trading and research.
I've used the research platform by a former CTO of Quantopian, J.Larkin at https://github.com/marketneutral/alphatools
and would recommend you look at it as a way to start your quest.
I can tell you where we are and what some of the pitfalls are if you email me at [email protected]
alan
The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian.
In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.