Yesterday morning I was talking to a VC about statzen when the topic of human perception of averages came up. It is a topic that I regularly bring up when talking about analytics. I regularly work from the premise that we are horrible at gauging averages over time. The VC seemed to agree completely and told me about some studies on the topic in regards to the stock market. He said that people tend to be overly optimistic and to weigh recent history disproportionately. I hope to find the study he was talking about so that I can learn more, but it did get me thinking about stock charts.
Below is an example of a typical stock chart. This chart is for ticker symbol FXE which is a currency fund tied to the value of 100 Euros in US Dollars. I am using it because it is one of several funds that I have been watching to keep an eye on the US economy (I also own some FXE shares). There is 1 year of historical data displayed in this chart.
As you can see from the chart, FXE has done quite well over the past year (this isn’t really good news for the Dollar, but that is another story). As I look at this chart thinking about the tendency to weigh recent data more heavily, I can’t help but think that part of the problem is that we are looking at historical data on a scale that doesn’t make much sense. The chart above shows a great fund that has grown pretty steadily.
I think that the 1 year scale makes it easy to look at this chart optimistically. Out of curiosity I wanted to see what a chart showing the daily percent change would look like. Using Google Spreadsheets (and the GoogleFinance function) I was able to quickly throw together a finance chart showing exactly that for the same 1 year time frame.
To me this chart tells a much different story. I think this chart does a better job at conveying the risk associated with a currency fund and counters the optimism of continued growth from the daily close price chart.
Here is a variation on that chart, but I smoothed it out with a 5 day average (I used 5 days instead of 7 because there is only data for days the market was open).
However, investing is really about the rate of return. I know I can get 4% from a low risk investment, so what would my cumulative 1 year percent change be? By looking at the cumulative daily percent change I am able to see what the rate is. As you can see, this is indeed a well performing fund. This chart mirrors the daily close price, but it makes the rate easier to quickly identify. Another cool thing about this chart is that if you start with the day you purchase an investment then the chart shows your actual rate of return. Using the same Google spreadsheet it would be easy to add the number of shares that you own to show not only the rate, but the actual dollar figure. (This chart looks flatter than the first chart because the data is displayed in less vertical space.)
None of these charts are a silver bullet to fix the problem of being too optimistic and weighing recent data too heavily, but it does make it a little easier to comprehend. The cool thing about the Google Spreadsheet is that you can fairly easily change the start date, end date, and ticker symbol and the charts update. I can see how someone who really knows what they are doing with stock analysis could create some spreadsheets with Google that could be of real value. I wonder if we will start to see people selling Google Spreadsheet templates the way there is a market for Excel templates.
Maybe next I will play with a chart that shows RTL and IYR ETFs (Exchange Traded Funds, kinda like an index fund) for Retail and Real Estate. It would be interesting to see how those two sectors compare. Maybe I should also add the price of oil in there to see if there is any correlation there.
You can view the spreadsheet that I used for this post.
