Simple Reversion Tendency Indicator?

Quote from dunleggin:

I use a model that spits out the type of 'good' mean-reversion picture that you seek, but from your (understandably) opaque description I'm not sure how well it may apply.

'Fundamental data' ...such as an econometric time series? Periodicity? What does 'short-term' mean in your strategy?

Are you able (willing) to upload a subset of the data, obviously disguising the actual nominal data value, in csv or xls format? I could quite easily run it through the model and graph the output.
More...

Attached is a sample of ~2000 bars of fundamental input, EMA(fundamental input), signal (just over/under vs. EMA) and closing price. My hold period is roughly 5 daily bars and I'm holding a large portfolio of overpriced/underpriced securities, so it's sort of statistical arbitrage where I just need a slight edge that can be replicated across many instruments to get a strong & consistent return. Positive signal = short, negative signal = long. In practice I would only open a position when the signal were strong enough, but this is the theory.

There's a lot of residual risk because you can be right in terms of relative pricing but on the wrong side of a bull/bear run. Hedging and diversification is critical.

But... I can increase my edge substantially if I can distinguish between the instruments that have a strong tendency to quickly revert to their mean from those that have historically stayed "mispriced" for long periods of time.
 
Quote from sf631:

But... I can increase my edge substantially if I can distinguish between the instruments that have a strong tendency to quickly revert to their mean from those that have historically stayed "mispriced" for long periods of time.
More...

The way to make money with mean reversion is to study and trade capitulation and euphoria.

Daily indicators alone will not do this for you, as they are lagging and mushy, and won't help you catch true turning points effectively. They are simply averages of action over an X period of time IN THE PAST, which is typically meaningless if you really want to get in on the action at the present time.

You can sometimes use mushy indicators for general setup and creating watch lists, but to trade mean reversion effectively you will need to dig into intraday periodicities and price action. That's where the profits are made.

(Example: See NXPI 5 min chart May 24.)
 
Quote from 1a2b3cppp:

How do you mean reversion guys deal with situations like this?

http://www.elitetrader.com/vb/showthread.php?s=&postid=2075844&highlight=trend+followers#post2075844
More...

I read that post and that's exactly what happened to me when I was sim trading CL for several months. I'd honed a RTM strategy where I faded moves to "extremes", which I defined as an overshoot of the 2.0 Keltner channel. I was making simbucks hand over fist and felt silly for sim trading all that time and not taking it live. So one day I start a position on a strong overshoot and price sort of hangs there, then takes off again and I add more, it hardly pauses and runs again and I decide now it's REALLY extreme so let's just start doubling down... One big ugly trade and I wipe out weeks of gains.

A little trend-following seed was planted that day. I'll tell you the nicest thing about trend following: The trend is your friend until it ends, and when it ends it offers early warning signs and a chance to switch sides.

On the chart you linked to above, if you were short, there was an upper trend line break (early warning), then a break of previous resistance, either of which was a valid signal to reverse long and feel the love that comes from being on the right side of a better-than-expected move :)
 
Quote from ronblack:

?
More...

The Casino has an undisputed edge. That they have to occasionally pay out large sums does not change that. My RTM edge is not nearly as solid as a Casino's, but I have an edge, and the occasional big loser does not change that. I once has a position go 150% against me, and that was not leveraged, just short. Closed that trade for a 70% loss. It's a tough way to trade, that's why it's profitable. I am paid to take risks that most don't want to take.
 
Quote from lindq:

The way to make money with mean reversion is to study and trade capitulation and euphoria.

...to trade mean reversion effectively you will need to dig into intraday periodicities and price action. That's where the profits are made.

(Example: See NXPI 5 min chart May 24.)
More...

lindq, was your entry price in NXPI on 5/24 calculated automatically or you used daily indicators to put it in your watchlist and just watched the charts looking for an entry? In hindsight, you can see clear capitulation in 5 minute bars, slope of the decline accelerating, volume after open going from 25-50K shares every five minutes to 50-230K after 10am. You could have also placed a support trendline around 26.40 on the daily charts.
 
Simple reversion will not hold up over time, especially for the higher frequency durations. In fact, the "shake" algorithm has been largely abandoned for pairs in stat arb as far as I can tell.

What rules are you going to use for a stop-loss ? That is the major consideration in my view.
 
You must log in or register to reply here.
Back
Top