Building a trading strategy. Need inspiration

[Aber]

Hi,

My school are having a Quant trading competition where need to come up a trading strategy. Sponsor will make sure that data will always be available. I have good "university" understanding of mathematical finance and I have good experience in solving statistical problems in R

However I find it difficult to apply my knowledge in a Trading Strategy that needs to perform when tested on Market data.
Can you make some suggestions? Name some strategies and I can look into.

My background:

Undergraduate in mathematical economics
Curreuntly master student in mathematical economics

Knowledge on mathematical finance more or less corresponding to the book by Bjork: Arbitrage Theory in Continious time

Advanced programming experience in R

Thorough understanding of probability and statistics

Understanding of Machine Learning ( ANN, Logistick Regression, EM, Clustering, Gaussian Mixture models, density estimation etc)

 

[daleholborow]

What range of time will your investment competition cover?

 

[Aber]

Hi. I will be investing around 5-7 hours per week until January 1st. 2018. But we are three in the group (same background as my self) and the others will probably spend a little less than me. We will all be investing roughly around 200 man hours in this project. Maybe more.

 

[yetanotherquant]

First of all mathematical finance (if you mean risk-neutral valuation of derivatives) has very little to do with trading (unless you try to detect arbitrage opportunities).

Secondly, some "classical" approaches are chart patterns (https://www.cis.upenn.edu/~mkearns/teaching/cis700/lo.pdf but to my experience it doesn't work [anymore]), trend following (it may work), mean reversion.

You may get some inspiration from my book
https://www.amazon.com/dp/3000465200/
(my track record: Somewhat better than DUCKS | wikifolio.com )

 

[Aber]

First of all mathematical finance (if you mean risk-neutral valuation of derivatives) has very little to do with trading (unless you try to detect arbitrage opportunities).

Secondly, some "classical" approaches are chart patterns (https://www.cis.upenn.edu/~mkearns/teaching/cis700/lo.pdf but to my experience it doesn't work [anymore]), trend following (it may work), mean reversion.

You may get some inspiration from my book
https://www.amazon.com/dp/3000465200/
(my track record: Somewhat better than DUCKS | wikifolio.com )

I mentioned Mathematical finance because I wanted the readers in here to have a good understanding of my academic proficiency. thanks for your input