Let's talk about Hurst and how I use it to make sense of this crazy market. We can define Hurst (H from now on) in three regimes.
1. H=0.5 Random price walk, very hard to predict. Do nothing is my go to move.
2. H<0.5 Mean reversion is present. Price is more predictable. When H<0.3 there is a change in trend (up or down) going to happen. I make my biggest bets here because I use @42macroDDale research to tell me what is the most likely outcome from a econom and markt reg. perspective.
3. H > 0.5 this is where the extention moves happen. When H is high (>0.8) on multiple time scales we can expect an extention move up or down in price. Personally I find these harder to trade, but if you bought the right stuff at 2. you can sell here.
Let's pick a real world example. Gold recently had a mean reversion signal H<0.5, although slim. On June 21 to June 24 it triggered. A change in trend happened here. After that H went back to around 0.7.
I highlighted the period Gold H<0.5 with green arrows.
Ofcourse this has to be combined with a deflation or inflation economic outlook and that's where DD comes in.
BTW picture stolen from
Parallax Financial Research, Inc
Have great week,
Tom
Oh and it works on crypto too :-). I present to you: the biggest trade I never made. Extreme low H around #BTC 10k. Whatever. I'm not mad. No really.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
@Oz25583407 You calculate H and RV. Project RV (daily range width) for the next day and scale by H. So for instance, if RV is 100 and H 0.8 you get 20 upside points and 80 down. Works for me.
@Oz25583407 Also, H can stay high for a while when a market is trending strongly... spotting the difference between a price to make sales or stay in the trade can be discovered be tracking a lot of assets.
@Oz25583407 Which can either confirm or deny your view. Tough part is what do you do when the signal is NOT clear... that is where risk management comes in imo.