Geometric return is one of three methods for calculating return over multiple time periods. It is commonly used in backward-looking performance reviews because it takes compounding into consideration. There are two versions: 'total geometric return' and 'average geometric return'. This is time-certain, meaning, the order of returns does matter.
Synonyms: geometric mean, geometric average, compounded rate of return
As an example, let's say an investment had a compounded return over two
periods at 5% each period. The total geometric return formula in Excel
=((1+0.05)^2)-1 or 10.25%.
In math and statistics, a distinction is often made between discrete and continuous data. Log return is the more theoretical continuous version. In the practical world, however, most people think of returns broken into discrete periods instead.
So geometric returns is a compounded version where you add one to the return, take that to the power of the number of periods, then at the end subtract one. See below for the average geometric return.
Because it compounds returns over time it is most resembles an investor's actual performance, so it is used in backward-looking performance contexts.
Leo: Professor, I'm soooo looking forward to
your lecture on
geometric return .
Doc: [silently] It's 4:25. What was he doing 5-minutes ago?
This video can be accessed in a new window or App here , at the YouTube Channel or within this page below.
Geometric return definition for investment modeling (4:54)
The script includes two sections where we visualize and demonstrate the calculation of geometric return.
We're sitting in Excel and this is a snippet from our boot camp course.
We'll simplify this to two-periods but it applies to 3, and beyond. Here we have two versions of geometric return.
The geometric total return, sometimes called geometric return, involves adding 1 to each return, mulitplying them together, then subtracing 1.
And for the average, often called the geometric mean, take that total to the power of one over the number of observations. That's what this caret symbol means, here we'd raise it to the 1/2 power. I put the total here in double parentheses because that's similar to the logic in Excel.
Let's now demonstrate the 'total' and 'average' calculations using two hypothetical stocks, over three periods.
ABC had returns of 10%, followed by -11%, then another positive 10%. XYZ had returns of 100%, followed by -50%, then it was flat.
Total geometric return is calculated by adding 1 to each return. You
could multiply them together manually, like this, and subtract one, or
=PRODUCT function, with the
range of returns in parentheses, minus 1.
We have observations here. And for the geometric average take that same product, raise it to the power of 1 over 3, then subtract 1, like this. Or you could use a cell reference to the cell containing the number of observations.
And to review, you would choose the geometric method for backward-looking perforamnce reviews because it measures the compounding of returns, and I put an example here showing the change in dollars, starting with $1,000 so you can follow the logic. Notice this, and this.
The sentence earlier about time-certain is meaningful now. Let me show you why. Arithmetic returns, are more forward-looking, when you don't know the order of returns, unlike here. It is more reflective of the estimates for future periods. Take a moment to think about this.
Using the same data, but using arithmetic returns, look at the differences. 2.5% versus 3%, 0% versus 16.67%. To see why follow the link at the end.
Click box for answer.
Arithmetic, because the order of returns doesn't matter
Still unclear on geometric return? Leave a question in the comments section on YouTube. Also, see a tutorial page and video on Stock return calculation methods from the Quant 101 Series on YouTube. There we go over when to use arithmetic, geometric and log returns.
Our trained humans found other terms in the category return math you may find helpful.
The FactorPad YouTube Channel is about learning. Please join us.