Hello,
can you please suggest the best method how to evaluate and find global minima for cost function?
My cost function consists of evaluating the portfolio performance (CAGR, Sharpe, Draw Down, Profit Factor etc.) and has 3 input parameters.
I tried to use scipy.optimize functions and also dlib package which works quite good, but I do not like I must pass the number of iterations as a parameter, which is subjective.
Hence I was wondering if there is any better way how to find global minima. I performed some googling and found genetic algorithms can be a good candidate for this but I do not want to hop on wrong path.
Any suggestion or advice appreciated
Hello Tomas,
An alternative to finding the global minima of the cost function can be seen here. Using iterations for finding the global minima is also a decent approach to find the same if the iterations is adequately large and the learning rate is comparatively small.
Hope this helps!