Time series forecasting with Deductor Academic 5.2 - page 3
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Not familiar with this kind of product yet.
It will be interesting to see.
Just wanted to ask one question for clarification:
These kinds of predictive systems work without regard to fundamental analysis.
Could this seriously affect (in a negative way) the quality of their forecasts?
Just wanted to ask one question for clarification:
These kinds of predictive systems work without regard to fundamental analysis.
Can't it seriously affect (in a negative way) the quality of their prediction?
There is no fundamentals, but there is the psychology of crowds ;)
It's just that the deductor analyses 10 times faster than MQL4, but it also needs to adjust a lot of parameters (number of history data, neurons, epochs, % errors.....),
>> so that the result takes into account the technical and foundation analysis.
Friends, good afternoon.
I apologise for the inconvenience.
Could you, when you have the time and opportunity, give me some guidance on how to work with Deductor?
Such complications arise:
Do I understand the concept correctly:
1. we use the s_forDED_MA script to generate a data file which we will import into the Deductor project.
2. In the Example_MA_H1 project, specify the file generated by the script in step 1 as the imported file.
specify a file to export data generated by the analysis in Deductor
3. Run script s_GraphFiles, which "draws" data from the export file.
________________________________________________________________________________________________________________
Am I getting the idea of the work right?
I just have a problem at step 3 - the data is not drawn on the graph... :(
Please advise how to fix it.
Thank you very much in advance.
There's just a problem in step 3 - the data doesn't show up on the graph... :(
1.Read here.
2.Post the export file.
To set your own value, open the .ded file in Notepad++, find the line <RecognitionErr>0.05</RecognitionErr>,
change 0.05 to your min.error value, save it, open it in the deductor, re-optimize it with the new value, see % of recognized samples....
The problems are about the same as with the optimizer's selection of indices in TS - until the trend changes everything is fine and sometimes even very accurate (forecasts), when the trend changes - the bugs... And increase of reversible forecasts, etc...
More powerful tools are in STATISTICA6-8 (the same principle - use data from CDS file, process-forecast-executed in CDS) the Russian site is full of information about working with the program... I use it and a couple of other programs ... + my vision ...
ps.it's a thankless job predicting prices....
But so eager to look into the future...)))))))))