I have a question that has probably came up before but the most similar thread I could find (about C2 accounting) appears to have been closed. The questions is this … how can a system make trades that exceed (sometimes considerably) the system account size shown on C2? For example (just picking one system for illustration) the VIXTrader system shows a trade opened on 1/26/17 for 2362 XIV at an average price of 64.65. This woud require about $153K, or at usual x2 margin for overnight holds, about $77K (the closed date is shown as 3/9/17). The account size shown on C2 now is about $45K, and was less on 1/26/17. Just curious how a system can make trades that require more overnight margin than exists in the account.
It was a cumulative number.
He did like bot 100 and sold 80 and bot 80 and sold 80 and bot 80 and sold 80 and so on. Those add to a total of 2362.
Click on “Show AotuTrade Data”, you will see.
-
VIXTrader strategy does not take any margin!. Basically it is holding on average less than 50% of the funds in the market.
-
Click here https://collective2.com/details/106901765 on “Show AotuTrade Data” to see the real numbers of shares. You can do the calculation and see that in any moment no more than 100% of the funds has been invested in any giving day.
-
This trade start on 1.23.2017 and closed on 1.26.2017, and not as you wrote in 3.9.2017.
Thanks wmwmw for your answer
Thanks Cat … that makes sense now. I’m downloading all of the trade record data to CVS and pulling it into a larger Excel file where I’m trying to build up a “portfolio of systems”, and some strategy combinations were exceeding the margin limits I had imposed and that example trade flagged a problem. I’ll have to see if I can download the autotrade data as well. It would be a lot easier to build a portfolio of systems if the subcription rates could be based on volumes traded (which is known from the autotrade data) with some lower limit for the economics to work. If someone had $1.7M to trade ROptions they are paying the same monthly fee as someone trading it with $100K.
Robert … yes, wmwmw explained what is happening, but the trade record does show a single entry with an open date of 1/26/17 and a close date of 3/9/17, and a total of 2362 shares. So that it what I was using as my example. There is only one entry showing an open date of 1/23 and closed date of 1/26, and that was for 527 shares (I’m just reading the posted C2 trade record data).
To complete the picture for you please see the trades under “Show AotuTrade Data” of my strategy https://collective2.com/details/106600099
Here you can see that on a rare occasions the strategy took margin.
Hope that this point is fully clear now
I was not “picking on” your system in any way. It has been 11 years since I had a C2 account so am getting back up to speed on how systems are presented, and my question was a general question about the account size vs. the size of trades showed in the trade record. I just chose your system as a random example to pose the question, but was not implying in any way that there was a problem. Now that it was explained that each entry in the trade record can be multiple trades that are each individually opened and closed separetely, it is all clear.
I just wish to add my broken record here:
A few of us were complaining for a while now about the bookkeeping of trades in C2. As the system developer of this specific system explained, in his system the discrepancy is rare. But in some systems it is so significant that it basically makes what you try to do (calculate real life margin requirement) impossible. You want to do this for portfolio testing, I want to do the same thing for checking the ‘smoothness’ of a system – the goal/motivation is irrelevant – the UI and the downloaded csv is not enough for this. (And it is extremely unlikely that C2 will change their ‘trade’ concept.)
Another, more involved solution is to download the raw data using C2Explorer. I am working on variants of this for a while now, but it is not a smooth sailing either. The environment doesn’t allow defining methods, using joins, so I end up exporting the results of my simplistic queries and do adult-level processing in Java or in Excel. Even with this, I keep running into data consistency issues that I try to fix manually or programmatically. Sometimes I can, sometimes I can not.
End of broken record.
Good luck!
Joseph