I have a degree in Economics with special interest in Finance and Investing, so apart from English and French, one of my main languages is math.
When I first learned about Investment, Risk Management, and Financial Engineering, I felt like a caveman being shown fire for the first time. The calculus, algebra, and derivatives were at a level that was way over my head. But the funny thing about education is this: first, you learn how to do it, and then you use computers that do it for you.
When I research a stock, I do so qualitatively and quantitatively. Qualitatively, I look at the non-numerical aspects of the company… How is the leadership? Is the company creating new and innovative ideas? How are they viewed in the eyes of the public? Are they loyal to their customers? Are their customers loyal to them?
Quantitatively, on the other hand, my research is MUCH more in-depth and complicated. It includes frequency distributions, fundamental analysis, predictive models, and a few valuation methods that can be applied according to different criteria the company may (or may not) meet.
And when I conduct my quantitative research, the first thing I look at is the beta.
What Is Beta?
Beta is a measure of sensitivity to changes in the market. It is a ratio; a division question. For example, if Stock XYZ increases 10% while the S&P500 increases 10%, the beta is said to be 1. If, however, the stock increases 20% while the market increases 10%, the beta is 2.
Another way of defining beta is in terms of volatility. When we talk about beta, we are comparing the volatility of your stock compared to the stock of some kind of benchmark or index, typically the S&P500, Nasdaq, Dow Jones, or TSX/S&P Composite.
Why Calculate Beta?
Beta is risk that cannot be reduced through diversification. It is an important number that I use in other calculations and valuations that I will write about in later posts, so it is important that it is done properly the first time… otherwise mistakes can be made. Expensive mistakes.
When I analyze a stock, I never listen to what others tell me. After all, have you ever selected a stock on the word of someone else, only to find that it didn’t work out? So when I see that a website has calculated the beta of a stock, I do not believe them; instead, I do it myself to make sure it is done right.
Math doesn’t lie. It can’t tell you if you should buy or sell, but it can help paint a picture of how a company – and therefore, how a stock – may perform in the future given certain conditions.
Here’s how I calculate the beta of a stock.
Gather Stock Data
The first thing I do is collect data from a reliable source. For my purposes, I use Yahoo Finance – it’s free, accurate, and easy to navigate. For this post, I’ve looked up the ticker symbol “ABC”.
To find the necessary data, click on “Historical Data”.
After clicking on Historical Data, you’ll see a series of rows and columns with dates, dollars, and volume amounts. These are previous opening and closing prices of this stock from earlier days, and you can go back as far as you like, so long as it includes the life of the company (you wouldn’t be able to look up historical prices for Facebook in November 1999 because it wasn’t around back then).
Now we have to change how we see the data. This part is tricky; there is no set rule on how far back you should look, or the frequency of prices you should be looking at. For example, looking at daily data over the past 6 months is probably pretty useless… that’s only 2 quarters, and the data will likely be very noisy!
Beta is a measure of sensitivity to changes in the market. It is the risk that cannot be eliminated through diversification.
However, looking at 10 years of data with a monthly frequency may not be useful either. Imagine looking at data between 2005 and 2015… our dataset would include the 2008 Financial Crisis, and would likely skew your data.
For myself, I use weekly data dating back at least 5 years, so long as there is no unusual market activity during that timeframe (like the 2001 Dot-com Bubble or the 2008 Financial Crisis). As long as you can rationally articulate why you chose any particular timeframe, you should be good to go.
For the purposes of this post, I will regress data from Jan 2015 – Jan 2020.
Next, click “Apply” to populate the table, and download the data into your favorite spreadsheet program. On to the next step!
Gather Market Data
Remember that beta is a ratio of movements in the stock over movements in the market. So, we have to repeat Step 1, but with an index.
I use either S&P500, Dow Jones, NASDAQ, or TSX, depending on what stock I’m researching. If you are researching a Canadian company, it may be best to compare it against the TSX; however, if the majority of their business is done in the USA, another index may be more appropriate. Use your discretion.
You must ensure that the time period of the market is the same as the time period of the stock!
For example, if you download weekly historical data of ABC prices between January 10th, 2010 and January 10th 2017, you will want to find the index data for the exact same timeframe! The dates must line up!!
Set up your timeframe, click on “Apply”, and download this data.
Open up the data in your spreadsheet program, and give yourself a pat on the back. Data collection is now complete! On to the next step.