SPX Time Cycles - A Teaching Moment
· No noticeable changes in the SPX cycle composite since the last post.
· But let me zoom in for a teaching moment.
· I often repeat to use cycle composite peaks and troughs as approximate turning points and not to infer price levels from the cycle composite amplitudes.
· What makes cycle composites approximate?
· Cycle periods, at least for those cycles considered to be stable and reliable, are fairly accurate; they could vary slowly over time around their stable periods, but that’s a tertiary effect.
· Cycle phases experience continual small phase jitter due to the algorithm constantly trying to determine the optimum phase, but that’s also a small effect.
· The main culprits are the cycle amplitudes. For some cycles, they could vary noticeably over time.
· The cycle detection algorithm estimates the average cycle amplitudes over several cycle periods.
· When amplitudes are constant, the estimates are very accurate, but when they are time-varying, the average may differ from the true present and future cycle amplitudes.
· That’s one of the main reasons why I frequently re-evaluate the cycle composite; that reduces the problem but does not eliminate it.
· To confirm the turning point estimates, I usually look at confluence of cycle peaks and troughs.
· For example, the figure below shows that local cycle peaks, past and future, in this short time period are produced when 8td and 16td cycles have approximately synchronous peaks.
· I also marked Troughs #1- #3 with black ovals.
· Through #1 includes throughs of 8td, 16td, and 119td cycles.
· Through #2 includes throughs of 8td, 16td, and 27td cycles.
· Through #3 includes throughs of 8td, 16td, 27td, and 119td cycles.
· Generally, a confluence of more cycle troughs will result in a more significant price low.
· By that logic, Trough #3 may be more significant and longer lasting than trough #2 because in addition to 8td, 16td, 27td cycle troughs it includes a longer 79td cycle trough.
· Also, when troughs of multiple cycles are more synchronized in time, rather than staggered, the price spike is typically more pronounced.
· Also, when some long cycles point up while others point down, they sort of negate each other's influence, and we pay more attention to short cycles.
· On the other hand, when several long cycles point in the same direction at the same time, the impact of short cycles is smaller or negligible.
