On August 12 this year a number of mid-cap biotech stocks spiked in price by 20-50%. This included the stock of Pacific Biosciences, a company I happen to follow, which increased in price from around $8 to over $12. I showed the chart to my colleagues that day and we speculated about what could’ve caused such crazy price action. A few days later, I was looking at the chart on Google Finance and in my brokerage, and I noticed that the reported maximum value for Pacific Biosciences on August 12 was around $8. This is a whole 50% lower than the true maximum value on that day. This observation also held true for several of the other stocks that spiked that day.
I was aware that the values on the public charts like those based on Yahoo Finance or Google are averaged from the transactions taking place within small windows of time on the scale of a few minutes. However I was unaware that the charts are retroactively averaged again to remove outlier windows that can be up to an hour long like the ones occurring on August 12. For some analyses this outlier averaging could have a considerable impact. Additionally it means that the maximum/minimum prices we read on the charts can be off by a considerable factor like the 50% I observed.
I found it fascinating to realize that I don’t understand how the charts I look at so often are generated and thought I’d share my realization. Does anyone know how the averaging works exactly?
Leave a Reply