5 replies to this topic

[airedale88]

using SPX cycle lows...... the last 3 samples of the nominal 10 wk cycle averaged 47.5 days. as of monday's close SPX is approx 43.5 days along. the last 3 samples of the nominal 5 wk cycle averaged 22.8 days. as of monday's close SPX is approx 22.5 days along. the last 3 samples of the nominal 2.5 wk cycle averaged 11.3 days. as of monday's close SPX is approx 9.5 days along. the last 3 samples of the nominal 6/7 day cycle averaged 6.6 days. as of monday's close SPX is approx 2 days along. aside from the 5 wk sample average all other cycles due to bottom at this 10 wk nest of cycle lows suggest a wed thru friday (16th thru 18th) time window.

[Tor]

So the shortest and longest cycles need more time?

[The End]

Only 3 samplings. Is that enough?

[airedale88]

to account for the minor variation in cycle length usually present Hurst suggested using the average of the three most current lengths to estimate the timing of the next cycle low. an example, the nominal 10 wk cycle has shown a range of 48 to 56 weekdays over the last 25+ yrs. over the last few yrs it has gradually worked it's way down to the 48 day or even shorter lengths from that historic range. limiting samples to the three most current should be sufficient for estimates.

Edited by airedale88, 15 May 2007 - 07:47 AM.

[MangeMan]

Is that weekdays as in calendar days, not actual tradingdays?

[LeroyB3]

Airedale counts all weekdays whether they are trading days or not, but not weekends. Best, LB