Can one approximate TE DD from CE DD and MAR?

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
fab1usa1
Roundtable Knight
Roundtable Knight
Posts: 383
Joined: Sat May 21, 2011 8:30 pm
Location: New York

Can one approximate TE DD from CE DD and MAR?

Post by fab1usa1 » Tue Mar 17, 2015 8:23 am

Does anyone know of a formula to calculate Total Equity Drawdown from Closed Equity Drawdown and MAR ratio? Just a ballpark estimate.

My gut tells me that it might look something like this:

Code: Select all

Total Equity Drawdown = Closed Equity Drawdown * (1 + 1 / (MAR * MAR))

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Re: Can one approximate TE DD from CE DD and MAR?

Post by sluggo » Tue Mar 17, 2015 9:13 am

I think there might be a couple of problems with your proposed formula.

Problem 1: MAR Ratio is (CAGR / Worst_Drawdown_Ever). And Worst_Drawdown_Ever might have occured several years ago. On the other hand, Closed Equity Drawdown and Total Equity Drawdown tell you how much pain you're suffering today. How can it possibly be relevant today, what happened in the subprime mortgage crisis of 2008 {if that was your worst drawdown ever}? How can it possibly be relevant today, what happened when the NASDAQ bubble burst in Y2000 {if that was your worst drawdown ever}? Why should we imagine that drawdown today is somehow related to drawdown made many years ago?

Problem 2: MAR Ratio is (CAGR / Worst_Drawdown_Ever). And CAGR is the long term slope of your equity curve. This includes profits and losses of trades you made many years ago, many all-time-equity-highs ago. Why should we imagine that drawdown today is somehow related to the profits and losses made years and years ago, long before the current all-time-equity-high?

I think another difficulty might be that people are divided into two opposing camps. Those who embrace "Closed Equity Drawdown" do so because they believe "paper losses" (also "paper profits") are insignificant and misleading. Wild gyrations while a trade is held, are immaterial; the only thing that matters is the entry price and the exit price. And so Closed Equity Drawdown carefully ignores paper losses and paper profits. In the other camp, the people who embrace "Total Equity Drawdown" do because they believe there's no such thing as a "paper loss" or a paper profit. Marked-to-market profits and losses are real to these people. If they're US futures traders, they actually pay income taxes on end-of-year "paper profits" even if the positions are closed out at a loss in the new year. A profit that you pay taxes on, is a real profit, they believe.

So it seems odd to calculate both of these numbers; people who want CEDD will disdain TEDD, and vice versa. Or so I imagine.

fab1usa1
Roundtable Knight
Roundtable Knight
Posts: 383
Joined: Sat May 21, 2011 8:30 pm
Location: New York

Re: Can one approximate TE DD from CE DD and MAR?

Post by fab1usa1 » Tue Mar 17, 2015 9:49 am

Thank you, Sluggo. Recently I've taken a liking to closed equity performance data.

I am evaluating a third-party swing trading equity system. The author publishes only closed equity drawdown and returns over a 10-year period from a database of 20,000 listed and delisted stocks. One performance table is the top 10 worst closed equity drawdowns. I am trying to get an idea of what that worst drawdown would be in terms of total equity.

When I first joined Trading Blox in 2011 there was a series of posts from you where you spoke about the tendency for MAR to go to 1 when you increase your backtesting data over longer periods of time. (Please accept my apologies if I remembered it incorrectly.) From my own experience I agree. I can take an excellent system and backtest it over a short period of time and obtain a wide range of MAR values depending on the start and end date of the simulation.

Regarding my formula, I was hoping to get the proportionality correct. It just seemed right to me, in a roughly approximate way, that a system with a MAR of 0.5 should have a total equity drawdown that is 5 times greater than the closed equity drawdown. And that a system with a MAR of 1.0 should have a TE DD that is approximately twice the CE DD.

Thanks for your input.

fab1usa1
Roundtable Knight
Roundtable Knight
Posts: 383
Joined: Sat May 21, 2011 8:30 pm
Location: New York

Re: Can one approximate TE DD from CE DD and MAR?

Post by fab1usa1 » Tue Mar 17, 2015 11:26 am

No need to reply, Sluggo. It's the alchemist in me just trying to turn lead into gold.

Post Reply