Stock Volatility

Replies to This Discussion

Permalink Reply by Simbarashe on June 21, 2010 at 4:28am

First, is it for 6.5years? I am not sure whether you want the calculations for 0.5years because from the figure 6,1/2 years I can't tell which is which. But in my opinion, the use of the proxy (group of comparable banks) is the way to go, if you don't have your own data for the corresponding period. However, you will still need to adjust those volatilities by a factor to account for the uniqueness of your bank.

One more question, are these volatilities meant for internal or external stakeholders? What is the weight of your bank in the basket of proxy banks?

• ▶ Reply

Permalink Reply by sergei on June 21, 2010 at 8:55am

I agree with the above opinion. You can use a proxy, but you also can compare the last 0.5 year performance of your bank with the peer group's performance and make appropriate adjustments if needed. However you should also make a note about the methodology when presenting your results to ultimate users.

I also wanna ask about the methods to forecast future volatilities of earnings, cash flows and other fundamentals. Time series for such measures of performance are usually short to use statistical methods such as ARCH/GARCH. Does anyone used to forecast fundamental's volatility?

• ▶ Reply

Permalink Reply by Hemal P Patel on June 21, 2010 at 9:02am

Hello,

Yes, i am using 6.5 years of daily closing price for calculation purpose. Here's what i did. I took natural log of today's closing price over yesterdays closing price for the 6.5 year period. I then took Stdev for all the natural logs and multiplied it by square root of 252 for the annualizing factor. I came up with a percentage of 42%. I am pretty comfortable with the concept via all the research i've done so far and with the definition that the stock, a year from now, will trade up or down 42% from the closing price today 2 out of 3 times. (please let me know if i am missing anything and if my understanding is on point). Now, my bank has only been public for 2 years so i have about 3 banks in my peer group with the same asset size as ours. They all have 6.5 years of stock data and i have also calculated the volatilites for each. Here's the methodolgy i've used.

Bank Yrs. Months. Days Volatility Weight Weighted Volatility
My Bank 2.4.11 43.19% 50% 21.59%
Peer Group Bank 1 6.6 40.48% 50% 20.24%
Peer Group Bank 2 6.6 28.92% 50% 14.46%
Peer Group Bank 3 6.6 50.41% 50% 25.21%
Peer Average 6.6 39.94% 50% 19.97%


I used 50% weight but i dont know if there's a basis for using a certain weight? if so, what would it be? should the weight be 10%, 40%, 60%, ...this, i have no idea...and should i add (+) my bank and the peer group bank's weighted volatlity together to come up with my bank's figure?

This is for internal stakeholders.

Let me know what you think! Thank you so much.

Hemal-

• ▶ Reply

Permalink Reply by Aaron Cother on June 21, 2010 at 12:58pm

You are seeking a standard deviation number for you stock. I'm not sure that identifying a 66% chance of a stock trading in an 84% window is of much actual or even theoretical use. You might have better results with moving averages vs actual stock prices.

I am guessing that you need a volatility number to do some type of option calculation - either for employee benefits or for a capital markets trade and need a good input. Since the bank is just now public for 2 years, i'm guessing employee benefits are being tuned - in which case, stock price is important. If it's a capital markets trade - you might consider looking at a more fundamental valuation approach.

One other post mentioned looking at interest income, treasury income...etc. I would take that one step further when you are identifying your comps, look for banks that have similar income/balance sheet math. Banks have been under so much pressure that you have very different characteristics between banks - healthy banks, failing banks, and healthy banks that have taken over failing banks. These are three very distinct buckets over the next 3-5 years.

• ▶ Reply

Permalink Reply by Hemal P Patel on June 21, 2010 at 2:05pm

Aaron,

When i am looking at moving averages vs. stock prices, should i look at 30 days, or 60 days, or 90 days, etc...? also, when i take the stdev of past 30 or 60 or 90 days for the entire 6.6 year period, should i then annualize it? I am not sure if i fully understand the moving averages calculation conecpt.

• ▶ Reply

Permalink Reply by Unni on June 21, 2010 at 10:59am

The last 2 years have not been good for banks, interest rates are distorted, business models have changed and hence any projections based on limited historical data may not be a good idea. It is better to have a larger population (say Bankex as a whole or segregated by large scale/medium scale/small scale depending on where your bank fits in).

Rather than looking at the stock prices, I would look at:

Interest income & interest spread for my bank vis-a-vis the universe
Fee income for my bank vis-a-vis the universe
Treasury income
Other income
Level of non-performing assets & consequent provisions

and then compare Book Values & P/Es of the various banks to get a better understanding of stock price movements.

Levels of borrowings/deposits by government, industry, private, as also inflation during the periods under review would have important bearings on bank performance and therefore stock prices. Interest rates prevailing in the domestic market, as compared to international markets would also affect bank performances.

Anyway, statistical extrapolations have limited value in these circumstances.

• ▶ Reply

Permalink Reply by Bernard Kiarie on June 22, 2010 at 1:01am

And how about using the beta concept?Select the beta of your most closely compared bank.Unlever it to remove the capital structure proportions.Then using the expected capital structure (debt and equity proportions) of your bank,you lever the beta and there you go!!You have the beta for your bank

• ▶ Reply

Permalink Reply by David Ptashne on June 22, 2010 at 10:40am

Hemal -

From my experience, it is customary to look at industry comps as a proxy for expected volatility. But remember, what you are (most likely) looking for is expected volatility, right? There is a lot of debate as to whether the past is the best predictor of the future, esp. now.

I would suggest looking at implied volatility baked into the price of options on comparable institutions, which you can back into based on readily-available option prices (i.e., see Bloomberg).

If you ask me, I think you are on the right track by analyzing peers and looking at volatililes over different time periods (when market conditions were different).

• ▶ Reply

Permalink Reply by Hemal P Patel on June 24, 2010 at 8:54am

I agree, the past two years of industry as an whole has been in a turmoil and hence, not a true picture of normal stock behavior. I still feel comforatble atleast 4 and half years of stock data for the peer group banks because of mirroring financials and performances. Although, only thing i am having a hard time figuring out is the the weighted average i should use for the peer banks. For example, i've used 50% of the peer bank volatility and 50% of my bank's volatility for hte appropriate periods. The "50%" is what my question is? what would be an ideal weight to use? and how so?

Hemal

• ▶ Reply