Recent content by eitanhir

first, scaling is not relevant when using correlation, since correlation subtracts the mean of every variable and divides by the std. second, i don't understand the log suggestion somebody gave you. either you're interested in the variables correlation or in their logs correlation. the...

i'm sure it can be used as a trading platform, but i think it would be a bit cumbersome, there are more suitable languages for that. i used it as a system development platform, but only because i'm familiar with it from work. obviously it's mathematical power is wasted when using it that way...

pca is good for reducing the number of dimensions. i don't think this is what you need (maybe i don't understand your situation). i think using kmeans is a better option - it would help you find clusters in your 12 dimensions space. each cluster can have different mean result.

if u want, u can send me the file u tried to load, i can take a look at it.

read this http://faculty.haas.berkeley.edu/odean/papers/Day%20Traders/Day%20Trade%20040330.pdf the abstract of this paper says it all. (80% losing money)

hi all, i'm trying to do short term (few days) swing trades on stocks. from your experience, what is the best exit strategy for these kind of trades? there's always the dilemma between giving your position enough room to develop and not leaving to much profit on the table. what do you think...