I run two analyses when I want to include longer cycle periods in the analysis. I start the SPX at 1994 in one case, and a second run from 1980. Start with these dates and run the ICM setting in Sentient to see what you get. You will get confusing results with Hurst’s nominal model. You should then consider the cycle periods produced by the ICM analysis, that is, input them as a nominal model and rerun ST. BTW, I don’t look at the RUT.
So why these start dates? It helps ST identify the ~7-7.5 yr cycle period that gives lows in 1980, 1987, 1994, 2002, 2009 and the most recent being 2016. The 45 month cycles (~2:1 harmonic) that go with that view are 1994, 1998, 2002, 2005, 2009, 2012, 2016. I listed the cycle periods I use as a nominal model further up this thread. There will be many that don’t use this interpretation and prefer Hurst nominal model. However William Randall’s bandpass filter work on the SPX gives the definitive view of these cycles and respective lows. This analysis is what I used to project the timing of the Jan/Feb. 2016 lows. Hurst’s nominal model would not have given you that projected low (a low perhaps, but not as a 7-7.5 yr cycle low).
If William has the time, it would be great to see an update of his 45 month (=Hurst 4.5 yr) cycle bandpass filter analysis. Once you study it, everything I suggest here will make more sense.