Search
× Search
Sunday, December 22, 2024

Archived Discussions

Recent member discussions

The Algorithmic Traders' Association prides itself on providing a forum for the publication and dissemination of its members' white papers, research, reflections, works in progress, and other contributions. Please Note that archive searches and some of our members' publications are reserved for members only, so please log in or sign up to gain the most from our members' contributions.

Improving computerized system performance

photo

 Ron Jaenisch, Author, Andrews and Babson Technical Analysis Expert

 Thursday, October 1, 2015

https://youtu.be/ciGgtBjv3g0


Print

7 comments on article "Improving computerized system performance"

photo

 Oleksandr Medviediev, Portfolio Analyst – FIBI

 Thursday, October 1, 2015



What your Signal is based on (assume you are using single settings for all pairs) ?

Other than that must disappoint you - it just happened that performance summary has been "improving" as you shorten test-period from 3- down to 1-year. It's not "TestForward", it's not "out-of-sample" - it's just a coincidence.

If you flip and extend test-period to opposite direction from 3- to 4- to 5-years -- overall performance will respectively deteriorate, correct?

Yet overall impression is positive, keep up with good work!


photo

 Ron Jaenisch, Author, Andrews and Babson Technical Analysis Expert

 Thursday, October 1, 2015



Thanks, in 2011 the daily ranges expanded dramaticly.we started with early data and you can see the recent walk forward. I was fortunate enough to have private conversations with Dr Alan Hall Andrews........its his tech that we include.


photo

 Ron Jaenisch, Author, Andrews and Babson Technical Analysis Expert

 Thursday, October 1, 2015



If all goes well we will get the funding to finish this project


photo

 Ron Jaenisch, Author, Andrews and Babson Technical Analysis Expert

 Thursday, October 1, 2015



walk forward tests are almost as useful as judging a woman by how she looks in a bikini. When you see her and get that tingling feeling all common sense has just left you. I use a T test and the score is 11.83.......if you want details just email me at gmail.com


photo

 Oleksandr Medviediev, Portfolio Analyst – FIBI

 Friday, October 2, 2015



Good metaphor worth a fortune, I wish you get an adequate funding to make it an ultimate success

=))

Yes, in fact i would like to PM, what's you gmail?


photo

 Ron Jaenisch, Author, Andrews and Babson Technical Analysis Expert

 Friday, October 2, 2015



add my name to gmail.com and you are there....i have a website using my name also


photo

 private private,

 Sunday, October 4, 2015



When setting up automatic trading systems based on historical data, a very strong effect is given by the calculation method of the used software and choiceof the back/forward sampling ranges. Also depending on the type of strategy, if your data is tick data or other. ...Last not least

* the simulated test under real market data feed conditions for at least some weeks will show, if due to slippage and other influences the program will produce profit or losses.

* Life market test trades with a small account can confirm the performance of the algorythms.

* When increasing the asset volumes, additional Stop Loss conditions like 0- x % below the last relative low of the Equity Curve can reduce the risks visibly.

* So, if the money and risk management as base of any strategy is considered, you just have to be careful with overdoin curve fitting

5) then the systematic results will become moderately positive and successful.

Please login or register to post comments.

TRADING FUTURES AND OPTIONS INVOLVES SUBSTANTIAL RISK OF LOSS AND IS NOT SUITABLE FOR ALL INVESTORS
Terms Of UsePrivacy StatementCopyright 2018 Algorithmic Traders Association