Parabolic Yawn Oscillator

Following on from my last couple of posts on the Parabolic Time Oscilator (PTO), I've been playing around with another variation on this theme. David Varadi introduced the PTO which looks at the length of time a market has been above or below is PSAR using a ranking to measure the stretch.

I realised that not everyone will be familiar with the PSAR. It comes as standard with most charting package, but here's a visual in any case:

I investigated this and found that the PTO is generally mean reverting – i.e. the longer the market has been above its PSAR, the more likely a reversal is to come. More specifically, it doesn't dramatically improve the ability to predict daily direction, what it does help you to do is get on the right side of a volatility expansion.

Time stretch – What about distance stretch?

David's PTO measure the time a market has been above or below the PSAR. I wondered if looking at the size of the difference between the PSAR and the current price could yield any results.

Judging by the results from my last investigation, I proposed the following hypothesis:

• The further the market is above the PSAR, the more stretched the uptrend is making a big reversal more likely.
• The further the market is below the PSRA, the more stretched the downtrend is making a big rebound more likely.

David Varadi already proposed something called the Parabolic Stretch Oscillator which measured the volatility units earned during the trade instead of days. I wanted to measure something more simple initially – the stretch between the PSAR and the current price.

Let's call this the Parabolic Yawn Oscillator (PYO) for want of any further inspiration. In effect I think when the distance between the PSAR and the price gets stretched, it will soon shut like a yawn – ok I think I've over stretched this analogy now.

It's relatively simple:

Up yawn stretch: For days with the price above the PSAR, it measures the difference between the price and the PSAR then takes a 50 period percent rank of that difference.

Down yawn stretch: For days with the price below PSAR, it measures the difference between the price and the PSAR then takes a 50 period percent rank of that difference.

The indicator then varies between 0 and 100.

Here are the scores on the doors:

Trade entered at the close. No frictions/ costs etc. Just top level analysis for now. From 2000 on the S&P 500 cash. Long only.

Up yawn:

Down yawn:

Below is a chart of astrategy that went long when down PYO is above 0.5 – i.e. the market is overstretched below the PSAR. As you can see, things got a bit funky around the time of the credit crunch, but overall it does a reasonable job of capturing the bulk of the gains since 2000 from less than half the number of days.

Summary

So it seems the hypothesis was correct. In general, the greater the difference between the PSAR and the current price, the more the market is stretched and likely to reverse. No doubt returns could be improved further with tinkering or used in combination with other things.