Blog - Investing Notes

February 18, 2008 - Robert Taylor's Peculiar Definitions of Up and Down

Reader Jeff Miller requested that we evaluate the market timing value of the "Xyber9 Market Forecasts", as developed and presented by Robert Taylor, CEO of Trend Corporation, Inc. The company's web site states that Robert Taylor "was nominated for the Nobel Prize in Economics in March of 2000 for his work in proving the stock market to be predictable and not random as previously thought." It further claims that the Xyber9 program, "an Aero-Space system identification program originally developed for NASA and their rocket program," provides "market forecast accuracy...unsurpassed by anyone." However, using Robert Taylor's trend definitions and daily highs and lows for the Standard and Poor's Depository Receipts Trust (SPY) over the period February 25, 2006 through February 15, 2008 (497 trading days), we find that:

Robert Taylor's guiding principle in analyzing financial markets is "Taylor's Law," which states: "The financial market's expansion and contraction is qualitatively in direct correlation to the increases and decreases in gravitational fluctuations experienced at the human level. The increases in market price are in direct response to decreases in gravitational forces; the decreases in market price are in direct response to the increases in gravitational forces."

Since he does not describe the gravimetric data he uses as inputs to the Xyber9 program, we cannot test this principle directly. Further, we cannot locate any peer research on this principle via the Social Sciences Research Network or broader web searches. We therefore consider only the predictions for the SPY in the weekly (short-term) Xyber9 Market Forecasts. These predictions imply the following peculiar definitions of uptrends and downtrends:

An uptrend means that the daily high on the end date is higher than the daily low on the start date.

A downtrend means that the daily low on the end date is is lower than the daily high on the start date.

The weekly Xyber9 forecasts are generally for one to several days of uptrend and/or downtrend. Each weekly forecast includes a SPY chart that shows whether the forecast proved accurate. Missing from the presentation of Xyber9 forecasts are benchmarks showing what forecast accuracy rate one should expect, independent of any forecasting ability, based on the peculiar trend definitions that Robert Taylor uses.

The following chart shows the frequencies with which SPY shows an uptrend and downtrend over the next one, two, three, four and five trading days during 2/25/06-2/15/08, based on the above trend definitions. The chart shows that missing short-term trend forecasts, up or down, is fairly difficult. For example, the likelihood that a forecast for a two-day uptrend (downtrend) is correct over this period is 88% (84%), independent of any timing ability.

Note that qualifying forecasts with "or" alternatives, as Robert Taylor sometimes does, boosts these percentages. For example, the likelihood that SPY shows an uptrend as defined for XYber9 after either three or four days during 2/25/06-2/15/08 is 86%, compared to 84% for three days alone and 81% for four days alone.

Such forecasts are, of course, not economically meaningful to investors/traders because one cannot reliably buy at daily lows or sell at daily highs.

In summary, Robert Taylor's accuracy rate probably derives not from forecasting ability but from defining targets that are very hard to miss. The accuracy rate seems high only if one ignores the peculiar way he defines trends.

Regarding Robert Taylor's March 2000 nomination for the Nobel Prize in Economics: "The names of the nominees and other information about the nominations cannot be revealed until 50 years later."

For reviews of a few other methods (as well as reviews of some books and information web sites), see Blog Synthesis: Reviews of Books and Web Sites.