An interesting occupation for those speculating on the ideal Fed communique for restoring the health of the market on Tuesday is to consider the distribution of how many times S&P futures touch unchanged on the day. It seems to happen an inordinate number of times, given the distribution of open changes and hour- to-hour or minute-by-minute changes. I wonder whether such a count might be the basis of a market indicator, and whether it might be useful for predicting such things as direction and volatility, and for setting the ideal levels of collars?

As food for thought, here's a graph of intraday prices over the past six days.

John Floyd comments:

Looking at the NDX versus Fed Funds in 1999-2001, the Fed took several months to react. In fact the Fed tightened by 50 basis points in May 2000 after the NDX had declined by roughly 30%. Not incidentally, that tightening marked the high in yields; the two-year note yield subsequently declined substantially for a variety of reasons. In fact, Fed comments remained hawkish pretty much through November 1999, with a tightening bias.

The Greenspan Fed of the time may have had a slightly different character than today's Bernanke-led Fed. But recent comments have shown no indication that the Fed is ready to react to the market action of the past three weeks. Furthermore, it could be argued that Bernanke's Fed may regard any action now to be premature, inclined as it is toward targeting inflation.

Adam Robinson comments:

The number of times the market becomes unchanged would seem to be positive correlated with "uncertainty" — viz., the greater the number of times the market registers "unchanged", the greater the market uncertainty.

I say that as a starting point, so Vic's conjecture is probably a fruitful basis of an indicator.

Also, although stock price movements are not randomly or normally distributed, of course, Vic's query reminds me of the classic drunken man walking away from a lamp post, how often after n steps does he return to the lamp post (here in 1 dimension along a numberline, either positive or negative movements).

Of course, stock prices do not move in equal increments, so the mathematics would need to be adjusted accordingly.





Speak your mind


Resources & Links