One notes that a little counting is always good.

Since 1996 

………………….Expectation next

………..Number…1 day…10 days

new lows  249     2.8       9

new highs 538 -1.0        1.3

No need for Mr. Zussman on this. And put Faber in a box with Spencer.

One defined a new high as any 40 day high based on closing prices and a new low based on any 40 day low based on closing prices. One notes that the Faber study quoting some colleague was on pre 2007 data. But one doubts that their results would have been different had they bothered to count rather than talk about the data base back to 1790.





Speak your mind

3 Comments so far

  1. vic on March 20, 2013 8:42 pm

    that’s 40 days lows and 40 day highs. Why do technical things like buy the best and sell the worst never work after they are published? is it selection bias to start, luck from the regressions effect, too many splits to start with, or the result of wise people telescoping and thereby dissipating the effect?But always beware of those who have in their hip ” data in their data bank from 1787″ to bak up their promotions, however bespoke. vic

  2. anonymous on March 21, 2013 9:32 am

    Vic, i always enjoy your studies. Might you please describe what criteria you are specifically counting from 1996?

    You results are understandable.
    Thank you.

  3. Dean on March 23, 2013 8:11 am


Resources & Links