Screening Low-Correlation Pairs
1 visualización (últimos 30 días)
Mostrar comentarios más antiguos
How are you fellows? Hope everyone is doing great.
I have a question, regarding screening correlation matrix, for low-correlation matrix. Issue at hand has 11 asset returns, and thus produce 11-by-11 correlation matrix. I would like to screen out pairs that has low-correlation. I dont really have any minimums for number of assets that I would like to pick, and I am planning to use them for equal-weighting so I don't need to do portopt function. Although, I did try portopt, using same expected returns for all the assets, but didn't quite work. I have put small piece of code, if someone can find why it hasn't worked. I was hoping to put AssetMin = 0.2, so an algorithm would weed out few assets, based on correlations, but did not work.
Anyone has any idea how do I screen out for pairs. I would need row index and column index as output.
pval = 100; nasset = 11; assetmin = 0.2; assetmax =.5; pcons = portcons('PortValue', pval, nasset, 'AssetLims',assetmin, assetmax, nasset); [rs, ret, wts] = portopt(exwin, wincov,[],[],pcons)
Thank you
0 comentarios
Respuestas (0)
Ver también
Categorías
Más información sobre Statistics and Machine Learning Toolbox en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!