Thanks Copernicus! Yes, ideally we would like to find something in the market that is deterministic so that we can exploit it over and over. Just like a deck of cards that is nearly all dealt and the remaining values are known to the card counter, ...

FXEZ's Quantitative Research In R

Yes that's a visually pleasing way to show a deterministic 1 dimensional random walk. It is deterministic because in advance you know there will be exactly 5 wins and 5 losses - you just don't know the order. So the ideal betting strategy would ...

FXEZ's Quantitative Research In R

Thanks for posting this jimsterk. You're right that reproducibility is important. Your code does reproduce what is in the article. Doing it this way assumes you know in advance what your expectancy is and as I previously pointed out this amounts to ...

FXEZ's Quantitative Research In R

Good point. Yes and after making the adjustment to only include historical results up to the last bar, I still see similar results. It's worth noting that in a real world application there will likely be more variance in the results so the ...

FXEZ's Quantitative Research In R

Yes showing the comparison to standard deviation was the point. Can your standard deviation be computed: 1) in advance 2) without any data 3) and can it be projected out into the future to contain the approximate percentage of trading action from a ...

FXEZ's Quantitative Research In R

Inspired by this nifty graphic on Random Walks as well as the curves posted by alphaomega, Copernicus, PipMeUp, ezcurrency, 9047 and possibly others, I decided to roll my own one dimensional random walks. I also decided to incorporate something that ...

FXEZ's Quantitative Research In R

Thanks for posting this paper. It's an interesting idea - similar to altering bet size based on counting cards in blackjack. However as you point out it is dangerous. In retrospect (after you run a system test on historical data), expectancy is ...

FXEZ's Quantitative Research In R

Thanks for the clarifications. I appreciate it. Your point about independence of the terms in the formula stuck with me but I had to prove it to myself via simulation (I'm not going to post the proof) looking for a loophole that never emerged. So ...

FXEZ's Quantitative Research In R

Many thanks for the additional insight and information on this important topic, Copernicus. It sometimes takes me multiple times with a subject before it sinks in, particularly when I'm not working with it every day.

FXEZ's Quantitative Research In R

Very impressive, PipMeUp! I had a feeling your math skills would come into play in deriving a solution. A question about your starting formula: It looks like the (1+x) is simply a representation of percent gain and (1-x) would be a percent loss from ...

FXEZ's Quantitative Research In R

Wow, I'm humbled by the great contributions. Nice work guys! It sounds like you're using sampling with replacement of the original system's trade outcomes as in a bootstrapping procedure? Can you talk a bit about the procedure you're using to sample ...

FXEZ's Quantitative Research In R

The following three charts represent three different systems. All are profitable. In your opinion, which one is the best of the three? All three systems started with the same amount of money: $1000, and used the exact same money management strategy: ...

FXEZ's Quantitative Research In R

Thanks for your comment, vg10. Are you saying compute the cointegration on the correlation coefficient series? If you are using two symbols you only have one correlation series. How would this work in practice?

FXEZ's Quantitative Research In R

ezcurrency: It's been a while since I've played with cointegrated baskets. It seemed like the best input to use was converting price moves to dollar moves so you have a common denominator (as you have done). Also being aware of the JPY crosses' 100 ...

FXEZ's Quantitative Research In R

Thanks for your post Tardigrade, I'm glad that you're here. You make an excellent point about random number generators and their limitations. Have you used hardware based random number generators? I wonder how easily these hardward RNGs integrate ...

FXEZ's Quantitative Research In R

Thanks for your post alphaomega - very comprehensive and among other things stresses cost of trade execution - something that is often overlooked! I hope you will continue to contribute your findings here. I really appreciate it.

FXEZ's Quantitative Research In R

Thanks for dropping by Copernicus! Hope you get something out of it!

FXEZ's Quantitative Research In R

That's a great idea 9047! I created: url for this thread and have started moving over some of the files. This should help us stay more organized and make it much easier to find stuff rather than scrolling through the thread. I also added the link ...

FXEZ's Quantitative Research In R

#3) POE Percent of Equity money management comparison using opposite signals, and combined equity curve: What if we took regular and opposite signals and traded two POE accounts? And what if we averaged these two POE curves into a single equity ...

FXEZ's Quantitative Research In R