Skip to content

Code base for the AMA-K project. Which is an online portfolio selection algorithm that aims to search for optimal portfolios at each time step via the usage of market history and pattern matching techniques.

Notifications You must be signed in to change notification settings

khanhphan1311/Prediction-Project

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

91 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Prediction-Project

Research Project Code

Code base for research project into stock market algorithms and their performance on various exchanges.

The CORN-based models implemented based on their research papers. CORN-K provides the crucial inspiration as the algorithm uses pattern-matching to produce strong returns. CORN-K can experience some issues if the correlation coefficients are extremely high (very close to one) hence we introduce a new model AMA-K to assist if this is the case. We understand that it may be preferable to use a uniform portfolio in cases where the market is very unique as it may represent a rapid change that has not occured in the past.

The AMA-K model is CORN-based and CORN inspired. AMA-K tries to integrate alternative methods for similarity through the use of Online-KMeans clustering. Furthermore, it uses multiple agents over different time spans.

AMA-K needs more back testing and analysis to further validate its results. The method provided here was part of a research project.

About

Code base for the AMA-K project. Which is an online portfolio selection algorithm that aims to search for optimal portfolios at each time step via the usage of market history and pattern matching techniques.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages