MetaStock Format

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
TrendMonkey
Roundtable Knight
Roundtable Knight
Posts: 154
Joined: Fri Apr 22, 2005 9:14 pm
Location: Vancouver, Canada

MetaStock Format

Post by TrendMonkey » Tue Oct 18, 2005 10:24 am

Does anyone know where one can find the details of MetaStock's file format? (Eg. the information required to write code to parse it). So many vendors (including TBB) support it, that I assume its readily available, but I couldn't find it on Google...

verec
Roundtable Knight
Roundtable Knight
Posts: 162
Joined: Mon Jun 09, 2003 7:04 pm
Location: London, UK
Contact:

Post by verec » Tue Oct 18, 2005 6:52 pm

You may find this useful.

TrendMonkey
Roundtable Knight
Roundtable Knight
Posts: 154
Joined: Fri Apr 22, 2005 9:14 pm
Location: Vancouver, Canada

Post by TrendMonkey » Tue Oct 18, 2005 6:54 pm

I do indeed. Awesome, thanks very much!

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Wed Oct 19, 2005 1:07 am

hello Tim, TBB uses CSI data. i wanted to check the Data files and found it incorrect. For instance:

Coffee Data : KC

19840522, 217.60, 218.20, 215.90, 217.35, 2818, 11362,198407, 154.59

only close is correct , because KC N84 on 05/22/1984 had:
Open= 154.75 High=155.45 Low= 153.15 and Close was 154.59

i am confused :?

damian
Roundtable Knight
Roundtable Knight
Posts: 815
Joined: Tue Apr 15, 2003 8:43 pm
Location: dusseldorf

Post by damian » Wed Oct 19, 2005 4:54 am

BARLI wrote: i am confused :?
So it would appear.

My confident guess is that you are looking at a KC continuous back adjusted contract (or an equivalent synthetic time series of linked contracts). So that answers why your OHLC are different to the actual contract. Understand?

Secondly, you are wrong in your conclusion, the time series close isn't correct (at least as per your test of correctness).

Your data file appears to contain the following fields, the first seven are generally industry default, the 8th is an optional user customisation. Check your ASCII file field settings.

date O H L C V OI C[actual]

Regarding the 8th field, you have correctly observed that the actual close of the then current contract in the synthetic series is 154.59. But the close of your synthetic KC series is 217.35.

I genuinely think that you should take some time to familiarise yourself with the absolute basics before using any of this 'confusing' data to make trading decisions.

Tim Arnold
Site Admin
Site Admin
Posts: 8500
Joined: Tue Apr 06, 2004 1:41 pm
Location: Boston, MA
Contact:

Post by Tim Arnold » Wed Oct 19, 2005 7:30 am

Hi Barli,

I can understand why you are confused if you have not worked with backadjusted futures contracts before. It's ok -- we all go through this in the beginning.

The data you are looking at is our Trading Blox sample data, which include the new 'unadjusted close' field available from CSI. The data series is D,O,H,L,C,V,OI,DM,UC, where UC is the unadjusted close.

So the OHLC for the data series is backadjusted, but the unadjustedClose, is, well, unadjusted.

The process of backadjusting can be simple, or quite complex. But the basic issue is that we need to string together a series of delivery months, going back in time, so this new single data series can be used to most accurately test a trading strategy.

There is quite a bit written on the topic here and elsewhere. Here is a link to get you started:

http://www.traders2traders.com/papers/b ... tracts.htm

Hope that helps,

Tim

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Wed Oct 19, 2005 1:50 pm

thanks a lot damian and Tim! as some say "the worst question is the question that is not asked". :D

Tim Arnold
Site Admin
Site Admin
Posts: 8500
Joined: Tue Apr 06, 2004 1:41 pm
Location: Boston, MA
Contact:

Post by Tim Arnold » Wed Oct 19, 2005 2:00 pm

I agree. People should feel free to ask questions.

No such thing as a bad question.

Post Reply