I have quite an interest in random processes and exactly how we can go about looking for patterns in seemingly random phenomena. One interesting topic I wanted to bring up today was the use of neural networks to predict future stock market trends. What I will present is some free C++ code you can install and then try for yourselves whereby you can employ your own neural networks (NN) on real market data using some wrappers to ROOT (http://root.cern.ch/drupal/)! The strategy is based on buying a certain stock, in this case it is actually trading with a benchmark index known as the SNP500 (http://us.spindices.com/indices/equity/sp-500). The code is an extension to a wonderful package called Hudson written by Alberto Giannetti, he offers a GPL licensed end of day back testing framework using C++ stl, boost, ta-libs and gnu-gsl libraries ( original code base is here http://code.google.com/p/hudson/ ) where you can load in various formats of csv file (Yahoo, Google etc). This extension allows one to perform a multivariate analysis (MVA) which can decipher linear and non-linear trends between input variables of a given dataset and then test its performance. Please bear in mind that this was written during my PhD studies and was solely for fun one weekend. There are clear areas that can be improved in terms of the C++ I just don’t have the time right now, feel free to email me or leave a comment if you like for ways to aid the project. My extension can be obtained from the following location
# Download the repo git co https://email@example.com/mattreid9956/hudson-mva.git cd hudson-mva ./build.sh cd example/MVA_SNP500 # open the run.sh script and try executing the lines one by one to see if they all work, let me know if you have trouble.
This week there is an accompanying pdf file where you can find the detail of a study employed on SNP500 data obtained from Yahoo Finance.
Anyway check out the pdf file if you find it remotely interesting give me a shout. It’s not “complete”, but close enough and I wanted to get it out there and see if anyone is interested in collaborating (I am trying to finish my PhD as well as several other things and don’t like leaving things on the back burner for too long, so here it is). Anyone up to date with particle physics techniques will immediately recognise the format here, anyone who is not will hopefully gain an insight into how some of these techniques can be applied in a toy model trading situation. A network such as this may not be valid indefinitely, its effectiveness would need continuous evaluation, since like I say in the paper, a market is not static and there will be periods of non-normal activity (government intervention for instance). What would be interesting is developing methods of when to turn on a trend following algorithm and when not to! Anyway, if you have any thoughts let me know!