Volatility [Part 3 of 3]: How to identify stop-loss using volatility

Featured

The first thing you need before you initiate any trade is to identify your stop-loss (SL) price. The SL is a price level beyond which you will not take any further losses. For instance, if you are to buy Nifty futures at 5200, you may identify 5100 as your stop-loss level; you will be risking 100 points on this particular trade. The moment Nifty falls below 5100, you exit the trade. The question is: How to identify the appropriate stop-loss level.

One approach used by many traders is to keep a standard pre-fixed percentage stop-loss. For example one could have a 2% stop-loss on every trade. So if you are to buy a stock at Rs.500, then your stop-loss price is Rs.490 and you risk Rs.10 on this trade. The problem with this approach lies in the rigidity of the practice. It does not account for the daily noise / volatility of the stock. As a result you could be right on the direction of the trade but could still hit a ‘stop-loss’. More often than not, you would regret keeping such tight stops.

An alternate and effective method to identify a stop-loss price is by estimating the stock’s volatility. Volatility accounts for the daily ‘expected’ fluctuation in the stock price. The advantage with this approach is that the daily noise of the stock is factored in. This literally means that we are eliminating the chance of placing an irrelevant stop-loss.  Volatility stop is strategic as it allows us to place a stop at the price point which is outside the normal expected volatility of the stock. Therefore a volatility SL gives us the required logical exit in case the trade goes against us.

Let’s understand the implementation of the volatility based SL with an example.

This is the chart of DLF. For the sake of our understanding, let’s assume we are bullish on the stock and plan to buy it at 185 with a target of 210, expected to be achieved in the next 5 trading sessions.

Step 1: We estimate the daily historical volatility of the stock. Take a look at this post to understand how to calculate a stock’s volatility. Recall, volatility of the stock is a nothing but the standard deviation of the stock prices. For DLF as of today, the daily historical volatility is about 3.01%.

Step 2: Convert the daily volatility into the volatility of the time period we are interested in. To do this, we multiply the daily volatility by the square root of time. In our example, our expected holding period is 5 days, hence the 5 day volatility is equal to 3.01*Sqrt(5).  This works out to be about 6.73%.

Step 3. Calculate the stop-loss price by subtracting 6.7% (5 day volatility) from the expected entry price. 185-(6.7% of 185) = 172.

Step 4 : Estimate the risk reward ratio to check if it falls within your expected guideline. In this example we are risking 13 Rupees for a gain of 25 Rupees (Risk to Reward ratio of 1:1.9) which seems like a reasonable bet.

Note : In case our expected holding period is 10 days, then the 10 day volatility would be 3.01*sqrt(10) so on and so forth.

Conclusion

Pre-fixed percentage stop-loss does not factor in the daily fluctuation of the stock prices. There is a very good chance that the trader places a premature stop-loss, well within the noise levels of the stock. This invariably leads to triggering the stop-loss first and then the target.

Volatility based stop-loss takes into account all the daily expected fluctuation in the stock prices. Hence if we use a stocks volatility to place our stop-loss, then we would be factoring in the noise component and customizing the stop-loss on a case by case basis.

Volatility [Part 2 of 3]: How to calculate volatility of stocks?

In the PART-1 of the series we illustrated what Standard Deviation (SD) is and how it represents volatility. Now let us see how we can calculate volatility of stocks or an index.

Step 1) Calculate the returns of the stock price.

Return = [ (Ending Price / Beginning Price) - 1 ]

For practical purposes and ease of calculation, this equation can be approximated by:

Return = LN [ Ending Price / Beginning Price ], where LN is Logarithm to Base “e”.

Step 2) Select the return series and call the ‘STDEV’ function on excel and you get the standard deviation of the stock.

The SD or Volatility works out to 1.3%. To calculate volatility we have used last one years data of daily closing price of Nifty. Therefore we call this “Historical Daily Volatility”.

Here’s something important that you need to know. Volatility is usually expressed in annualized terms. So in order to convert daily volatility to annual volatility you need to multiply the daily volatility with the square root of time.

Annualized Volatility = 1.3*Sqrt(365). The annual volatility works out to 25.4%. Here’s a snapshot of the volatility information on ‘Nifty Futures’ as published by NSE. 

In the next post, part 3 of the Volatility series, we will discuss the effectiveness of using volatility to place stop loss orders.

Volatility [Part 1 of 3]: Is it the up & down movement?

This three part post attempts to throw some light on the following topics involving volatility

  1. What is Volatility?
  2. How do we calculate Volatility?
  3. An application of Volatility – Volatility based stop loss

What is Volatility?

‘It’s the up & down moment in the stock market’ – this is the typical answer you get from many market participants.  If you have a similar answer in mind, then we encourage you to read this post.

Investopedia defines volatility as – A statistical measure of the dispersion of returns for a given security or market index. Volatility can either be measured by using the standard deviation or variance between returns from that same security or market index. Commonly higher the standard deviation, higher is the risk”.

To understand the definition better, let’s look at a simple example. Take the case of two batsmen and the runs they have scored in 6 consecutive matches.

If you were to choose one of the two batsmen for the 7th match who is most likely to score at least 20 runs, which one would you choose?  Let’s do some quick math and get some basic stats on these players. We are going to calculate three variables namely,

1)      Sigma = The total number of runs scored

2)      Mean = Average runs scored per match

3)      Standard Deviation (SD)

Most of you will know what the first two variables are since its pretty straight forward, hence let’s just deal with what exactly standard deviations (SD) is. Standard Deviation is nothing but the expected variation from the average score.  In order to calculate the standard deviation, we first need to calculate the Variance (Var).

Batsman A has an average of 21.6 and his score in the first match was 20. Therefore his deviation from average was -1.6 for match 1! Similarly, for match 2 it is +1.33.

In order to calculate the Variance, we simply have to calculate these deviations from average, square them up, add them up and divide it over the total number of observations  - total number of matches in this example. SD is nothing but the square root of the variance. Here’s a picture that explains how exactly SD has been calculated.

So, lets go back to the stats of both the batsmen.

B seems to a better player based on ‘total runs scored’ and the ‘average score per match’! But if you pause for a moment and look at the standard deviation, then B seems to be an inconsistent player. His SD is much higher, indicating that the batsman is erratic and risky.

Lets actually put SD into some use and see what the number has to tell. One way to use SD is to make a projection on how many runs player A and B are likely to score in the next match. To get this projected score, you simply add and subtract the SD from their average.

B has a very wide range between 9.9 to 34.4. Because he has a wide range it is difficult to estimate what his likely score is going to be in the next match. He can either score 10 or 34 or anything in between with all the possible outcomes having equal probabilities. However A seems to be more consistent. His range is smaller which means he will neither be a big hitter nor a lousy player. He is expected to be a consistent and is likely to score anywhere between 19 and 23.

Now, going back to our original question, which player do you think is more likely to score at least 20? By now, the answer must be clear; its player A.

So in principal, we assessed the riskiness of these players by using “Standard Deviation”. Hence SD must represent ‘Risk’. In finance, we define ‘Volatility’ as the riskiness of an asset which is measured by standard deviation.

If you know the market’s volatility and the current market price then you can estimate the likely range in which the market is likely to trade. Let’s illustrate this with a quick example:

  • Nifty CMP = 5000; Nifty Volatility = 1.2%
  • The Lower Range for Nifty is = [5000 – (1.2% of 5000)] = 4940
  • The Upper Range for Nifty is = [5000 – (1.2% of 5000)] = 5060

If the volatility increases from 1.2% to 1.8% we end with a much wider range (4910 and 5090). Now assume you are long on Nifty at 5000 with a 50 point target. In both the cases (Volatility of 1.2% and 1.8%) there is a likely hood that your target of 5050 will be achieved, However if the trade were to go against you and you do not have a stop loss, you will lose more money when the volatility is higher. Hence higher the volatility, higher is the risk.