Page 1 of 1

Trend following and range trading

Posted: June 5th, 2020, 9:32 pm
by complyorexplain
We are preparing a paper for an economics journal and co-writer has questioned terminology. 

We are saying that it is "familiar to trend-following traders who buy on rises and sell on dips, that such a trend-following strategy will gradually lose money if the price remains within a range". Will most readers (and reviewers of course) understand the concept of a "trend-following" trader, and "trend-following" strategy, and of the idea of a price remaining "within a range"?

There is a subtle point that a price will always trade within a range (if we define 'range' as the values in between the high and low points) and of course a trend-following trader cannot be following a trend if there is no trend.

Re: Trend following and range trading

Posted: June 5th, 2020, 9:45 pm
by bearish
I would tend to say yes, but "economics journal" gives me pause. To be on the safe side, I'd suggest to just slightly reformulate your statement above into a workable definition of a trend-following strategy in your context. In particular, time horizon matters. Your reference to "the price" implies that your trading/observation frequency is sufficiently low as to make microstructure issues irrelevant. It's tempting to replace the idea of the price remaining within a range with a notion of mean reversion, which should be well understood, but they are clearly not exactly the same. 

Re: Trend following and range trading

Posted: June 5th, 2020, 10:26 pm
by Alan
There are interesting related subtleties in connection with option strategy replication by pure dynamic trading of the underlying. I am thinking of both portfolio insurance and the old paper by Carr and Jarrow:  "The stop-loss start-gain paradox and option valuation". 

Basically, suppose  the underlying dynamics is pure random walk (allowing a drift), and you can trade without frictions. Then, if "trend following" is thought of as a type of systematic portfolio insurance, you will *still* lose money (over time) -- relative to buy and hold. This should be true even if every arbitrary upper barrier is eventually crossed (as should be true with a random walk with a positive drift). Indeed, the (relative) money lost is the cost of the implied put option(s).

Re: Trend following and range trading

Posted: June 6th, 2020, 8:55 am
by complyorexplain
Thanks for these comments (and particularly the reference to Jarrow and Carr).

We formalise the idea of 'trend following' and 'range trading' within a paragraph or two, the sentence quoted is in the introductory paragraph. Co-writer is a gifted economist who has published widely in the field of monetary economics and (hence?) had not come across the term 'trend follower' or the informal idea of a price 'remaining in a range'. 

Of course there is no point in appealing to an intuition if the intuition is not shared by the average reader, but then the question is who the average reader (and average referee) is going to be.

Re: Trend following and range trading

Posted: June 8th, 2020, 1:23 pm
by Alan
There's a joke about "Lake Wobegon" here for those who know the name Garrison Keillor.