i love this thread ,very interesting
i will remember
i will remember
Algorithmic/automated trading and emotions 13 replies
Learn Computer and Data Science with Algorithmic Trading 5 replies
Dislikedhttp://etfhq.com/blog/ Backtesting results of many indicators, cross over systems etc.Ignored
Disliked{quote} Make everything as simple as possible, but not simpler (A. Einstein) Sometimes the simplest possible is already quite complex. We have to do with it.Ignored
QuoteDislikedThis PDF version is made available for personal use. The copyright in all material rests with the author (Simo S¨arkk¨a). Commercial
reproduction is prohibited, except as authorised by the author and publisher.
Disliked{quote} Much easier to read than Bayesian Forecasting and Dynamic Models by West and Harrison. Sadly, here also, the author mainly focuses on the Gaussian distribution. Also he doesn't talk about discounting factor for the unknown covariances (whereas West does). If we use the return instead of the log(return) we can use simple linear filters (without Taylor development). But the distribution isn't normal and MCMC is not acurate at all: look at the figure 7.4 page 128 and keep in mind the pendulum model is static while the trends in the market change....Ignored
Disliked{quote} Did you see the Matlab code working out some of the chapter exercises from the book? http://becs.aalto.fi/~ssarkka/ See Publications / Books for the publisher's website. I was thinking of converting some of the code to R where it is easier for me to use in a trading application from my platform. If anyone would like to work on conversion, or general collaboration regarding the text / code, please PM me. Chapter 13 seems like an interesting summary of possible methods, with pluses and minuses of each discussed.Ignored