There are some environments that are conducive to the kind of trading we do here at Quantitrader and there some that are like kryptonite to Superman.
Low volatility is one of those environments for Quantitrader. Low volatility is QiT’s kryptonite.
The best environment for us is high volatility like we saw back in 2008. That year was so good to QiT’s portfolios, I’ve chosen to remove 2008 from the performance stats posted on the website because it was skewing them too much.
I’m not going to explain volatility, and how we measure it, for there are hundreds, if not thousands, of articles that do a better job than I ever could. As well, I’m not entirely sure I agree with how we measure volatility but since I don’t have a better way I’m stuck with the VIX. With that said, the VIX, the instrument we’ve use to measure volatility since 1990, spiked above 80 during the 2008 financial crisis.
There is nothing we can do about the low volatility but there is something we can look at to see if what the portfolios are experiencing is cause for concern and if we should take action. I have the circuit breaker in place to ensure a drawdown does not turn into a system failure. However, there is a way to look at the portfolios from a different perspective called the “Rolling CAR.”
So what is a “Rolling CAR?”
If a trading system states last year it had a one year return of 36% it means if you started trading on January 1st by December 31st, you would’ve earned a 36% return. This is very simple and straightforward.
Now if this system advertised a 36% annualized return over six years, this would mean, if you started trading on January 1st, and quit on December 31st exactly six years later, you earned the equivalent of 36% a year. However, throughout those six years, any one 12-month period the system could have made only 1% – 2% and a different 12-month period it may have gone up 70%.
So a more realistic view of the CAR is what is called a “Rolling CAR” (as opposed to a parked CAR?). It provides a more realistic way of looking at a system’s returns
Rolling CARs are used to look at the returns of any system for the holding periods similar to those actually experienced by investors because it breaks a performance period into many smaller, overlapping periods, kind of like slowing a movie down to study it frame by frame. Most annual returns shown in system results assume you start trading on January 1st of that year and go through December 31st. By looking at a rolling CAR you can see the return for each one year period (or any period you choose) regardless of when you started during the year.
If your backtest period was January 1, 2010, to December 8, 2017, a Rolling CAR calculates your annual return January 1st, 2010 then “rolls” forward and calculates an annual return starting January 2nd, 2010, then rolls forward and calculates an annual return starting January 3rd, 2010, then rolls forward and calculates an annual return from January 4th, etc. Thus the term “Rolling.”
On the graphs below, the annual CAR is plotted for each day. Each data point is the return you would have made if you had started trading exactly 12 months prior (or any period you choose). For example on the Momentum Retirement Portfolio Rolling CAR chart below, if you had started to trade 1 year prior to 3/24/2014, your account would have made a whopping 74% CAR. If you had started to trade 1 year prior to 5/19/2016 your account would be down -12%.
So, rolling returns, because they draw on so many more data points than trailing returns, can reveal times when the trading system is in trouble or not.
The Rolling CAR charts do not install with Amibroker so I had to get them from my system programmer.
I would like to say I am using data back 2010, which is comprised of backtest data and real time. The real time data used is from the date of launch and is, of course, different for every portfolio.
This chart may a little daunting at first blush but let’s take it a little at a time. For these charts, I have used a benchmark CAR of 15% to illustrate a level I would consider the portfolio on “probation.” Not a time when I would quit trading, for it is still making money, but it needs to be watched. If the Rolling CAR falls below 0 and goes negative, it is time to take action and possibly invoke the circuit breaker. Remember though the circuit breaker is based on a different indicator so the Rolling CAR chart is used as another tool for verification.
Let’s take a look at the Momentum Retirement Portfolio’s i year Rolling CAR. If you had started trading this portfolio 1 year ago your account would be up 9.88%.
On the other hand, looking at the 6-month rolling CAR, if you started trading this portfolio 6 months ago, your account would be up 52%
If you had started trading this portfolio 1 year prior, your account would be up 27%.
If you had started trading Momentum Margin 6-months prior your account would be up 26%
If you had started trading this portfolio 1 year prior, your account would be up 29%
If you had started trading this portfolio 6 months prior your account would be up 54%
If you had started trading this portfolio 1 year prior your account would be up 10%
If you had started trading this portfolio 6 months prior your account would be up 7%
If you had started trading this portfolio 1 year prior your account would be up 3%
If you had started trading this portfolio 6 months prior your account would be down -11%
This has to be the biggest disappoint of 2017. It started its downturn as soon as I launched it.
If you had started trading this portfolio 1 year prior your account would be down -8%
If you had started trading this portfolio 6 months prior your account would be down -9%
Rolling CAR helps to see longer term in more detail because it forces us to look at the mountains and valleys we have to endure along the way to the overall CAR. If you see the ups and downs along the way you’re likely to keep a cooler head in times of market volatility, or lack thereof. Looking at a Rolling CAR gives you the confidence in a trading system to allow it time to reap the benefits. Basing a decision to sell on a recent plunge, for example, is what sent many people fleeing equities in 2008-09 when they should have been getting in.
Check our out the Performance Matrix for our algorithms and sign up today.