Conjugate Gradient Project Ideas

[squaremeplz]

Homework Statement

Hi all,

I am trying to think of some intermediate (C,Java, Matlab) project ideas that would use the conjugate gradient, least squares regression (basiaclly a lot of statistical inference methods), lasso regression, and neural network feedback in order for a computer to solve a problem.

The only problem I can think of at this time is image/pixel analysis but this has been done many times and I just want to come up with a quick simulation of my own. Could you guys throw some realistic project ideas at me for I lack in the creative department.

the main formula of course is 1./2 .*x^T.*A.*x = x.^T.*b

Any input is appreciated.

 

[lippyka]

Homework Equations1./2 .*x^T.*A.*x = x.^T.*bThe Attempt at a SolutionOne project idea could be to develop a program that uses the conjugate gradient, least squares regression, lasso regression, and neural networks to predict stock prices. The program would take in data about the stock market such as closing prices, volume traded, etc., and would use the formula given to make predictions about where the stock price will go in the future. The program could also provide advice on when to buy and sell based on the predictions it makes.