From recollection, Carol Osler talked about round numbers in one of her papers.

Testing High Leverage Compounding

Based on my understanding, the general data mining bias test being discussed here compares a particular system or systems on real data to systems produced on random data. So it is essentially a test of actual historical results vs. results on random ...

Machine Learning with algoTraderJo

KaBo, You might start with this wiki page on Model Selection.

Machine Learning with algoTraderJo

Hi sym, I'm not sure how big my background is but I remember you also from another thread, and it is good to see you doing such innovative work here! I have looked at the options data in the past but didn't have success in generating alpha. I like ...

WWW power (daily COT)

It is interesting to note that the data that was created on Wednesday, available in the afternoon on Thursday, is used to trade on Friday.

WWW power (daily COT)

bistreevseh, Switch the slashes in your path from "" to "/" as R expects the latter. I've used 64 bit R with mt4r.dll without problems. The actual calls to Rterm.exe are programmatically made through the command prompt so it shouldn't matter whether ...

MQL4 -> R-Project - Interface Library

ezcurrency, It appears that you are using price for your regression and allowing an intercept. I found that either this type of model (price + intercept) or the price differences with no intercept models were more stable and logically sound than the ...

Trend basket by linear regression of uncorrelated pairs

Hi ezcurrency, I don't know how I missed this thread. Interesting work you've done here. I've done something similar in the past but it appears that your preprocessing step is superior to what I did to cull the list of pairs down by correlation. ...

Trend basket by linear regression of uncorrelated pairs

By the way, those of you who are using this project to access R might wish the R session to not shut down on an error, which seems to be the default R setting. You could fix it to always disregard errors by doing something like this (untested) edit ...

MQL4 -> R-Project - Interface Library

@roel13 The premise of the thread is that there is a single best prediction method. This contradicts a statistical learning theory known as the "no free lunch theorem." This theorem essentially states that there is no universal best method for all ...

Best Predictive Algo?

Hi roel13, Yes it is possible to weight formulas. But it sounds like you're still trying to decide what to do. Basically if you are dealing with boolean conditions: true/false you can use a simple voting scheme where 1 means your indicator indicates ...

How to add weighted conditions in trading formula

Thanks for your question Trenderer. It bugged me that I didn't know the answer so I spent some time on the taking partial profits subject. I still don't have a complete answer but I think I can say that there are some limited settings of R, r and ...

Expectancy Management

algoTraderJo, I'll try a simpler to understand explanation that hopefully expands the discussion a bit and puts us over the finish line on the ensemble divination issue. If you have an equity curve, how do you get the position series? A formula for ...

Machine Learning with algoTraderJo

algoTraderJo, Thank you for posting this followup. I have moved my analysis into R and now attempt to show graphically the result from my assumptions, flawed as they were. I'm hoping that this will show my reasoning and I can determine where the ...

Machine Learning with algoTraderJo

algoTraderJo, I see you have given us a homework assignment. Thank you teacher! I used Excel for this analysis. If I sum the two cumulative return series I get the following in terms of return and max dd: Column MaxDD LastPnL PnL/DD NN1: 0.3152 ...

Machine Learning with algoTraderJo

Thanks. I didn't come up with anything so I cheated by looking on Wikipedia: So it appears to be "diversity" but there is some question regarding how to measure it. After following one of the footnotes to a paper it appears that finding the "best" ...

Machine Learning with algoTraderJo

Yes I've seen that phenomenon as well as the inverse. When system results are worse as a result of a combination, it is sort of like an inverse Parrondo's Paradox. I guess the answer as to why is that it depends on the exact rules you are using for ...

Machine Learning with algoTraderJo

As someone who has been there and "built" a working framework, the points raised by algoTraderJo are very valid. My framework can't do everything listed in aTJ's first bullet point, but it accomplishes what I consider the most important parts of ...

Machine Learning with algoTraderJo

Hey KV427, thanks for the thread! I tried solving the matrix by hand and after a bit of effort was successful. But I'm really rusty with matrix math, and it's too easy to make errors with the calculation, and now my brain hurts so I moved into R. ...

Assessing the Profitability of All Possible Trading Strategies

For the developers, you might find this attached document useful concerning some of the ways to directly access R.exe / Rterm.exe in batch mode. Working With R in batch Mode: Some notes for beginners by Michael Lundholm ...

MQL4 -> R-Project - Interface Library