Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Recently the volatility is catching up and you already mentioned it has a major influence on the strategy (like many others).
Is it not better to choose the new parameters for the next period based on an optimization period that has the same volatility as now?
In other words ATR is around 0.02 now. Would it be correct to optimize from may till now to find the new parameters starting in January 2015?
Maybe better to use an optimization period starting at 10/4/2013?
Please your feedback from a statistical and experiential point of view
When I do walkforward optimization, the reoptimization times are set in advance. So, for instance, the reopt may be every 1 year. In that case, I will redo the opt 1 year from the last time I did it.
I have never tried reoptimizing based on the method you suggest. I think if you adequately tested such method, it might yield good results. But for me it would be a radical change.
Was just flabbergasting if you optimize on a period that had unusual behavior and at the same time is based on something that makes the strategy perform bad (low volatility), this will probably end up with parameters for the next period that will not do very well.... iff volatility is back to "normal" in the next period...
On the other hand who can predict the future and maybe volatility will be bad next year too....
When reading your book i was wondering through what process and reasoning you went to come to the 105 minute chart for the euronight strategy. Did you initially optimize on different time frames first or look at different timeframes to find the patterns or.....
I'd have to go back and look at my notes, but I believe that the 105 came about primarily because I wanted equal size bars through the night session. It did not come about from trying a bunch of different timeframes, and then picking the best one (that would have been optimizing).
Hi Kevin,
Would you have any recommendation on historical market data?
I am half way through your book and start realizing the importance of good backtest data.
I'll be happy with 1mins charts for CL, ES, NQ, E6 (and maybe a few others) pair dating back 5 to 10 years.
Backtest Data sell a future package for $1300 but it's tick data and contain way more than I need.
Any thoughts on cheaper or better alternatives would be appreciated.
I keep getting requests for people to share their GomRecorder data. A few threads have been started on the subject but not many are following thru it would seem.
I am collecting data from IQFeed via QCollector. All data contains bid/ask and is tick …
Hi everyone. I really enjoy reading this forum and wanted to share with you my collection of ES tickdata (from 2003 - June 2009). They are data from the CME and therefore should be pretty accurate. To download, please go to post#7 , Bigmike has very kindly …
I would recommend that for any data you buy, purchase, download, etc. --- make sure the data source you backtest with is the same data you trade live with. Otherwise, you run the risk of data differences leading to invalid results. A tick or 2 here and there could make a big difference. And then, if your use a continuous contract, different rollover dates can have a huge impact.
You are on the right path though with questioning data. Too many people use data without a second thought, and that can be dangerous.
Hi there,
I am trying to build a good set of historical data.
I have been directed by Mike to this thread.
So far I did the following :
Downloaded some Gom data (Gom-CL-Jul12.7z as a test)
Installed the Gom base indicator 2.6
Watched big mike tutorial …
We are actually two traders with with the exact same issue.
I know you aren't using NT anymore but if by any chance you could spot something obvious that would be awesome.