Trend Trading

Quoting synicz

Disliked

{quote} Although I am not here to cause an academic debate. I am really curious if you have evidence that currencies have greater tendencies to revert to mean compared to stocks. The reason why I asked is because there have been studies done on currencies and found that like other asset classes such as stocks and commodities, there exists exploitable anomaly. Which is trends or what some call momentum. Whether the returns are still profitable after deducting transaction costs and swaps is not yet known to my knowledge.

Ignored

Quoting synicz

Disliked

{quote} Although I am not here to cause an academic debate. I am really curious if you have evidence that currencies have greater tendencies to revert to mean compared to stocks. The reason why I asked is because there have been studies done on currencies and found that like other asset classes such as stocks and commodities, there exists exploitable anomaly. Which is trends or what some call momentum. Whether the returns are still profitable after deducting transaction costs and swaps is not yet known to my knowledge.

Ignored

Quoting Copernicus

Disliked

{quote} I am not so sure either. Here is a test of 45 instruments over 17 years across asset classes using the same fully automated simple daily momentum breakout model using a Donchian entry (150 period) and Donchian Exit (75 period) using ATR for an initial stop and position sizing. Performance results include spread, swap and slippage. Of the 45 instruments I have ranked their performance below. Look where some of the forex pairs sit in this context. {image} {image}

Ignored

Quoting lemna

Disliked

Forex swaps can also be a problem in holding positions long term, as you mentioned. I personally believe its an unfair tactic used by brokers as they definitely charge more than they have to, and very rarely offer positive swaps. In any case, there's nothing to be done, you just have to work with it. Let's just say for trend traders, its a bigger problem than the spread. Also, brokers can change these rates day by day, so once you are in a position, you just have to suck it up or exit. That doesn't mean it makes it impossible, its just something...

Ignored

Quoting Graviton

Disliked

I'll be dealing with some health problems over the next week or so. Hope to return when it's over. Best success to all.

Ignored

Quoting lemna

Disliked

{quote} Copernicus, that's really interesting. Any chance you have results for stock indices long only with the same test to hand?

Ignored

Quoting Graviton

Disliked

I'll be dealing with some health problems over the next week or so. Hope to return when it's over. Best success to all.

Ignored

Quoting Graviton

Disliked

I'll be dealing with some health problems over the next week or so. Hope to return when it's over. Best success to all.

Ignored

Quoting Copernicus

Disliked

{quote} Here you go lemna. A long only version of same universe (just refer to the equity indices in the mix) :-) {image} {image}

Ignored

Quoting lemna

Disliked

{quote} If I understand your stats right, the stop starts as 3X ATR 75, then is trailed by the 75 Donchian when it gets higher than the initial stop?

Ignored

Quoting lemna

Disliked

{quote} Thanks Copernicus. That is also very interesting and I am happy to be proved wrong! I have come to my beliefs about trends through using some different testing parameters. I take weekly data and go long if price is above X or X,Y lookbacks, short if vice versa. Just measure the results by percentage change, no adjustment for volatility. My thinking is that the equity curves you see from this are somehow purer, more representative of propensity to trend. Though I am not much of a statistician so I dont make any claims for my tests. See, when...

Ignored

Quoting synicz

Disliked

{quote} To be fair, what Copernicus shows is generally only to his 150, 75 donchain rule. Furthermore we do not know the YoY volatility. A stable yearly return may be better than a volatile high absolute return. If you know what i mean. @Copernicus Is it possible to show the yearly performance of these assets? As it will show us a better picture of the stability of the returns.

Ignored

Quoting Copernicus

Disliked

{quote} The findings are consistent across an array of perhaps 30 or so different short/medium and long term trend trading strategies I have tested over the 17.4 year period ranging from simple traditional trend following techniques to the more exotic variants. Consolidating monthly returns across 45 separate large excel files is a time consuming exercise to present the data. I humbly beg you to reconsider your request. Perhaps another day when I am feeling chipper about the world. :-)

Ignored

Quoting synicz

Disliked

{quote} Hey man no worries bout that! Just for the sake of furthering the discussion. An unrelated question, what software did you use for this analysis? Im currently using r studio for such backtesting and they always give me ugly, unreadable formats and displays!

Ignored