What is Sharpe Ratio.
How it is calculated. [LINKSYSTEM_67510989]
The Sharpe Ratio is the inverse of the coefficient of variation.
See wikipedia for details: http://en.wikipedia.org/wiki/Coefficient_of_variation
Mr. Sharpe gave the number his name, so his name is mentioned all over the place…
In simple words is a risk vs return stat. I don’t care much for it to be honest with you. It tells you if the higher return was due to higher risks, the debate is what is considered risk. I don’t think there is such thing as risk free asset, but is used in this formula.
In General a ratio of 1 or better is considered good, 2 and better is very good, and 3 and better is considered excellent.
Thanks,
Julio
A good example why I don’t like the Sharpe Ratio:
Sharpe ratio 3.081
Strawberry Rhubarb FX
Really 3?
I did find a good article with a good explanation:
A ratio was developed by Nobel Laureate Bill Sharpe to measure risk-adjusted return of an investment. It is calculated by subtracting the risk-free rate from the rate of return for a portfolio and dividing the result by the standard deviation of the portfolio returns. Or,
Sharpe Ratio = Excess return over risk free return/ Annualized standard deviation of returns
The Sharpe ratio tells us whether the returns of a portfolio are due to smart investment decisions or a result of excess investor risk. This measurement is very useful because although one portfolio or fund can reap higher returns than its peers, it is only a good investment if those higher returns do not come with too much additional risk. The greater a portfolio’s Sharpe ratio, the better its risk-adjusted performance has been. A lower number is worse.
Stocks have performed better, on a risk-adjusted basis than Treasury Bonds, because the Sharpe Ratio on stocks is higher than Treasury Bonds. A negative Sharpe Ratio is considered very bad. It means you could have done better, on a risk adjusted basis, by holding cash. The point of risk adjusted return is not to look at return in a vacuum, but rather to consider how much risk you had to take in order to generate “excess return” - the amount of return over a market benchmark or the risk free rate i.e. 10 year Treasuries or 90 day T-bill.
Mr. Martinez,
I don’t get your explanation. How do you propose we calculate SR for a system? We all can read the good article. You have asked your question: "Really 3?"
What is your answer? If it is not 3, then what is it ? Please enlighten us, if you can.
Brian,
I didn’t mean to imply the calculations are wrong. I’m just not a fan of the Sharpe ratio anywhere not just in C2. I’m not questioning why such system has 3, I’m just showing an example of a system with a Sharpe Ratio of 3 and what it looks like, so other users could get a feel for it. I’m sure there are users that like the Sharpe ratio and use it to make investment decisions, I don’t.
It is what it is, if the system has 3 then 3 it is. No one is going to argue that. I’m just saying, look at a system with 3 and the result, is up to the investor to interpret the number in their own way.
Thanks,
Julio
What result, Mr. Martinez?
Is it an unavoidable dd that you are pointing at?
Do you have your favorite ratio that you can share with us?
An interesting topic. I would like to share my 2 cents.
I do not like Sharpe ratio either. This is the ratio that brought down Long Term Capital Management in 1998 in a spectacular collapse. I think Calmar ratio is a much better measure for the systems on C2. Personally, I would screen systems using the following criteria:
1. CAGR beats S&P, and greater than 30%
2. Calmar ratio greater than 3 (Max DD at least 3 times less than CAGR, the rational is that the system would have three chances to recover within a year)
3. Max DD less than 30% (This is the line in the sand. Any system that crosses this line is a gambler to me no matter how high a CAGR is).
These criteria are quite straightforward. The caveat is that the CAGR showing in C2 is hypothetical for any system younger than one year, and you need to assume that the CAGR can hold its ground longer term. You need to assume the Max DD can hold its ground as well, but this may not be a valid assumption for many highly leveraged systems without proper risk management (no stops and position sizing). I would not want to believe anyone who tells me he or she uses mental stops, because these mental stops can fail easily sooner or later in the heat of real battles in a volatile market (thinking about the flash crash in 2010).
Based on the above criteria, the FX system that many people reviewed in the past few days is a good one to me. The vendor apparently uses hard stop, and the max DD is still reasonable (below 30%) compared to its CAGR. The only problem is its relatively short history. We should have better idea in 6 months.
Good luck and happy trading!
//TQT
[LINKSYSTEM_69417480]
I agree with you about max DD, it is less than 22% for the system.
And Calmar ratio is better than SR.
The problem with all such ratios is that they require years of performance data to be meaningful.
Thus we are left with GAGR and max DD for now, until more data is accumulated.
Happy trading to all!
I also like the Calmar ratio in conjunction with other risk measurements. On important measurement often overlooked is the win percentage compared to the average profit and loss ratio. Almost all successful long lived systems, whether stock, futures or forex systems have a low, sometimes even less than 50%, win percentage but a high avg. profit/loss ratio. "Cut you losses short and let you winners run" usually wins in the long run.
The forex system you refer to, apparently Strawberry Rhubarb FX, has just the opposite relation. This together with the very high risk and leverage the system takes, which is reflected in the high returns of the early months, spells trouble and I think the system will fail unless the developer changes his strategy.
Another high risk and high leverage forex system, even not as extreme as the afore mentioned, is Venters AUD/NZD. What bothers me is that when you read the systems description it almost sounds like the developer thinks he has discovered the holy grail and we all know such a thing does not exist.
"if the system has 3 then 3 it is."
Not really. C2 doesn’t include costs (commissions, autotrade fees, subscription fees) in its performance calcs. That makes all the calcs pretty worthless IMO but it’s a C2 problem, not a problem with a properly calculated Sharpe ratio.
Edit: I’m surprised the NFA hasn’t busted C2 for this.
Dennis,
My assumption is that C2 does include costs into the calculations. This is an educated guess based on them including these costs on the P/L and using the P/L to make their calculations. C2 would have to verify this. Whether it is calculated properly or not, I don’t use Sharpe ratio for deciding which system to trade.
Thanks,
Julio
You can tell they don’t include costs because changing the commission plan changes the monthly returns and the graphs but the stats and the model account status don’t change. Also the equity in the account status is the line on the graph with no costs.
