Posts tagged ‘Bootstrapping’
Monte Carlo Resampling of Equity Curves Using N-Bar Segments (paper)
Very readable paper on MC resampling techniques. Classic MC techniques tend to generate smoother equity curves with shallower drawdowns than actual live results, as they “chop up” returns too finely, thereby reducing the impact of correlation during Black Swan events. The paper presents a simple method which attempts to preserve these correlations (click on title to view paper).
Bootstrapping: White’s Reality Check
Interesting article on WRC (click title):
“Prior to WRC, bootstrapping could be used to generate the sampling distribution to test the significance of a single rule. White’s innovation, for which he was granted a patent, allows the bootstrap to be applied to the best rule found by data mining. Specifically, WRC permits the data miner to develop the sampling distribution for the best of N-rules, where N is the number of rules tested, under the assumption that all of the rules have expected returns of zero. In other words, WRC generates the sampling distribution to test the null hypothesis that all the rules examined during data mining have expected returns of zero.”
The Journey Begins
(This post is a continuation of “The Fateful Lunch”)
My head was spinning as I walked to the car after my conversation with James; The idea of using one’s PC to analyze and trade the market in a completely automated fashion was very appealing to me.
Why? Because Automated Trading not only provides a path to grow one’s wealth; It also presents your inner geek with the opportunity to delve into aspects of Artificial Intelligence and Bio-Inspired Algorithms (neural networks, genetic algorithms, genetic programming, reinforcement learning, fuzzy clustering, vector quantization, etc). Put more plainly: You get to play with cool stuff while earning some nice coin. It doesn’t get any better than that!