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)
On your example Excel formulas you used two different days held. One used 30 days the other 1/12 or 30.4 days.
I have used XIRR for years to compute my account's monthly and yearly ROI. I haven't used it for trades but you are correct it would be more accurate.
Here is what I use. B44 is End Date. B14 is Start Date
=(1+XIRR(G14:G44,B14:B44))^((B44-B14)/365)-1
This gives a result of 2.381% for your example on the Excel sheet.
My old formula =365/(B2-B1)/12*(B6-B8-B7-B9)/(B5*B4) result is 2.416%
Your new formula =+((B4*B5+B6-B8-B9)/(B4*B5+B7))^(365/12/B3)-1 result is 2.414%
It depends if you want to say that the day you acquire the option or the day you exit the option is a day held. I have been subtracting the start date from the end date so that only counts one of those. For example, March 31st minus March 1st is 30.
There are fees when you exit a position. NFA fees and maybe clearing fees. Broker Commission is either half in half out or all in. At Gain my entry and exit fees for ES are 0.56 per contract each way.
So I back-tested your strategy. I'm assuming you developed it by looking for a method that had a very high win rate over the back-testing period and that still produces good returns.
Where MM is the margin multiple of the initial SPAN margin and exit point is what the premium needs to be as a percent of the original premium to close the position.
Here are some stats:
Where,
Win Rate = % of scenarios that have a positive return
E(mROI) = average mROI
Med(mROI) = median of mROI
E | Loss = average mROI for losses
E(DTS) = average days from acquisition to liquidation
Med(DTS) = average days from acquisition to liquidation
Max(%M) = Maximum margin requirement as % of margin held
The scenario that failed is shown below:
Date: 12/1/2015
Short(-1): ESH6,ESH6,1660
Long(2): ESH6,ESH6,1420
Total Margin: 1,597.20
mROI for this trade is -13.4%
Does this look right to you? Do you get the same results?
I agree with all of your numbers for that trade except that trade didn't fail. You don't go on margin call until your maintenance margin is at 100+%. Initial margin is 110% of Maintenance Margin. So a margin call wouldn't happen until % of Margin was 110% or higher.
Did you backtest for the years 2013 through 2017?
Starting a trade on each trading day of all those years?
How long did it take?
How many lines in your database?
You downloaded each ES and EW3 option for each trading day for all of those years?
What parameters to pick which options to download?
Which database are you using?
Where have you been all of these years?
Your work could be a huge help in finding strategies. Thank you again.
Ron, The 1/12 in the XIRR formula is to convert it to a monthly rate not to a rate over the period of time the transaction was open. XIRR always returns an annual ROI, so raising the XIRR to the power of 1/12 is equivalent to the 365/12 adjustment made in the formulas.
Your adjustment to XIRR --> =(1+XIRR(G14:G44,B14:B44))^((B44-B14)/365)-1 is correct if you want to know what the ROI was from the start date to the end date; but if you want the equivalent monthly ROI, you would always use: (1+XIRR(G14:G44,B14:B44))^(1/12)-1
As others have mentioned this is incredible work. I do have one question, rather than writing out the trade to margin call is you could run this test and find out what the percentage of winners and losers are if you have a 200% and a 300% of premium exit point? I'm looking forward to seeing the results.
I ran 2x and 3x initial premium exit points, more losers and less overall ROI. The size of the losers is not necessarily lower either (on a monthly compounded basis). For example, the 2x premium exit has #NUM because it loses over 12% in 1 day; compound that for 30 days in a month and you have almost no money left.
Note that I assume you get out at the settlement price for a day and not at exactly 2x the initial premium. This makes sense to me; expecting to get out at a specific price when markets tank is unrealistic.
The first run is Ron's methodology. The second is 2x and the third is 3x.