Yesterday’s blog entry reported on an e-mail that I sent to Academic Researcher Wade Pfau on March 1, 2011. He sent his response later that day.
Wade said that he shared my view that using 30-year rolling time-periods to make the case for Valuation-Informed Indexing was “compelling” and that “I’m definitely going to be using tables like the one I showed at Bogleheads with all the risk measures.” He explained that: “My idea is to show many different tables with results over the whole period for returns and risks. Valuation-Informed Indexing always provides more returns for often less risk.” Wade’s e-mail continued: “I also want to read more about the “information ratio” measure, as I think it may be the most important of all the measures in that table. But I need to better explain its full power.”
He stated: ” I will pick some of the cases where VII did least well and show the 30-year rolling periods for those. This is to help avoid data mining. But the fact will remain that no matter what I try, VII will still perform better in 85-95% of cases for 30 years, and the times that it does worse are those in which buy and hold was doing its best in history (which is always right around the peak of the bubble in 2000), and so VII is not more risky. I have a new figure for showing this as well. And a nice figure showing the outperformance percentages across rolling periods of lengths between 1 and 40 years. I think it is all quite persuasive.”