Tester Issue.
ubzen,
Sorry for the delay in responding. I have been doing more in-depth research on this. I have found some very informative posts regarding this. As a result, I am now testing the quality/accuracy of the data download. For others who may be experiencing the same issues, I found the following extremely helpful:
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
MT4 Veterans,
I am having problems with tying out expected trades with tester trades. To improve the quality of my testing, I downloaded tick data from a broker in CSV format to over-ride the MT4 tick generation process. As a part of this process I utilized a script which took the CSV file and converted the data into FXT and HST formatted files. I then took these files and replaced the history residing my my history directory as well as placed the FXT file in my tester / history directory. In the FXT file I changed the attributes to read only so that the data within would not change.
I then took one of my EAs and ran it through the tester. Within the EA, I made a function to capture all of my variable values and take the data to a text file. I subsequently imported the text file into MS Excel and used formulas to replicate the EA logic. What I discovered is that the expected trades identified in the Excel file did not match up with the tester trades. In other words trades were taking place at different times than what was computed in the Excel spreadsheet.
My understanding is that when FXT data exists in the tester / history directory, the tester uses this data instead of computing ticks by way of the program's tick generation process when the tester is modeling "Every Tick".
By the way, after running the EA through the tester, I am getting 99% test accuracy.
Does anyone have any insight into what may be causing my discrepancies?