Volatility, Nifty and analyzing your system

Trading strategies are like seasons. When the season is in full swing, the person who follows a strategy that correlates with it feels like a king and the guy who follows a strategy that is antithestis of the that feels like shit. Currently, its a season of Mid and Small caps. Stocks in those pockets have gained substantially, better than what the overall markets have done for themselves.

When the markets are in a range, trend followers wonder where the good old days of smooth trends on rocks went by while mean reverters seem to be like god as they sell at tops and buy at bottoms. And then the season changes and every attempt at catching the top and bottom just goes for a toss. 

I have been in recent times investigating how volatility can affect system returns and while there is no conclusion as such that I am willing to lay upon the table, I hereby present some charts which may hold a clue to what kind of markets the system does best in.

To start off, the 1st chart on the table is one where I have taken the Geometric Mean of the daily range (High – Low) and then calculated the Standard Deviation of the same period

Image

Its no surprise to see 2000, 2008 and 2009 being the top years in terms of both range and deviation from the mean. After all, we saw maximum volatility during these years. What is interesting for me is 2012 wherein Nifty gained 27% while Nifty saw the least amount of moves on a intra-day scale. 

The Weekly Range 

Image

and the Monthly Range

Image

do not show any major difference compared to the daily range (other than percentage change). The way to use the above dataset will be to see if there is any marked difference in system returns in years of high volatility vs. years where volatility was low. Depending upon the average holding period for the system, one can compare and contrast with one of the above charts to try and see what kind of markets appeal to the system and what doesn’t.

The reason to compare in my opinion would be to figure out whether bad periods in the past (system) were due to some factor that can be in hindsight seen as a problem area for the system (you may actually have to break-down the above chart on monthly scale since its unlike systems to under-perform for years unless the logic itself is hugely faulty.

For example, check this Equity Plot (Plot starts in Jan 2012 and is upto 15/04/2014)

Image

The system had a excellent run in the initial few months (Jan – Feb 2012 being a runaway bull market) and then more or less settled down until it had one fiery run (this taking place from late July 2013 to Early October 2013). In this period, it was not a run-away market, but one where you could see huge swings (Wide Range Bars being many in the months of August and September). The system seemed to be able to take advantage of the volatility. But once the volatility ended, even though markets have climbed steadily higher from that point onwards, the system has returned zero returns indicating that the system got caught with low volatile periods and lost money (more or less – Since October 15, Nifty is up by around 800 points while the system in the same time has actually lost around 400 points) . 

Good systems (which have sustained profitable periods) are tough if not impossible to create. Markets being dynamic, no matter what one does, markets always seem to be able to get the better of us. Its hence important that one constantly revises his views and strategies while at the same time keeping in mind the fact that even the best systems can have a long period of draw-down due to unfavorable market conditions. As they say, do not throw the baby with the bathwater.

 

 

Advertisements

About Prashanth

Have been a full time participant in the stock markets since 1996. Run a Yahoo Group where focus is exclusively on discussions of the Indian Markets using Technical Analysis as the tool (groups.yahoo.com/group/technical-investor)
This entry was posted in Uncategorized and tagged , , , . Bookmark the permalink.

One Response to Volatility, Nifty and analyzing your system

  1. Nice write up.Yes Volitality and swing size affects the performance of trend following system.
    Hope the above Equity curve based on trend following based on moving average 🙂

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s