the social scientist.

Indicator Update

You are probably familiar with Fosback's high/low Logic Index. He took the number of weekly 52-week highs and lows on the NYSE (all issues traded I think), and calculated the lowest of the highs or lows as a percentage of total issues traded. If the 10 week average of these percentages fell below 1% that was a bullish signal. It didn't matter if the low number was highs or lows. At the beginning of a bull market the percentage of stocks making new highs is extremely low, and when the bull market lifts off, the percentage of stocks making new 52 week lows is very low. He wasn't publishing in 2002; so I don't how the indicator did in forecasting the bottom of that bear.

I have calculated the indicator for myself since 2004, and it is getting close to a bull signal, although I suspect it is wise to wait until there are very few new lows.

If the 10 week average is above 5% that is a bearish signal. The Hindenburg omen seems to be a variant of Fosback's indicator. The sell signal failed at the end of 2005, but forewarned of weakness in May 2006. It issued a sell on 8/17/07 at the intermediate bottom. It has not come to signaling a buy until this month. As of Friday the 10 week average was at 1.62% and falling.