Different Algos for Different Currency Pairs

[afeudale]

Sorry for the confusion Leon, I'm not up to date on the latest "thresholds" used for these trading systems, but it doesn't really matter.

The whole point is that the thresholds are different depending on currency pair. To say that 2.0 (or 0.8 or 0.2, etc) is some magic number that all currencies obey, in my opinion, is shortsighted.

Think of it like rolling a bowling ball up a mountain. If given enough strength, the ball crosses the top and trend/breakout occurs. Without enough strength, the ball comes close to the top, but falls back down (retrace). All currencies are like differently weighted bowling balls.

Furthermore, there are mountains of various "lengths" representing the different time scales. For example, USDJPY may need a monthly momentum (CSS) of 0.2 PLUS a daily momentum (CSS) of 0.4 PLUS an hourly momentum (CSS) of 2.0 in order for it to be a true trend. All others are "fake trends" and will reverse and/or have lesser chance of success.

Furthermore (again), these mountains are in a country with varying climate throughout the year and during the cold months, it takes more momentum to push it up the mountain and in the summer months, takes less momentum (simplified analogy of course).

Hope that helps.

Afeudale

You are continuously referring to the level cross value of 0.8 or 0.2 etc. Baluda's superslope has always been set with a level cross value of 2.00 and at times we have used 0.0 but I have never seen anyone use a difference of 0.80 in all my years of testing. Can you please be specific or are you referring to the threshold. Right now you are confusing the heck out of me.

Cheers
Leon

[Wavegarrick]

Thanks for the comprehensive answer Afeudale. Hope we can all learn more from this.

Cheers
Leon

[BostonEJ]

I think the use of CSS or CCFp varies depending on strategies. I see continued reference to trends where my interest, for now, lay more with reversals or mean reversion.

[afeudale]

I tried and tried with mean reversion for most of the last 2 years, thinking it was easier to detect and code than trend, but I found the contrary was true. I thought everyone was trend trading and I wanted to go against the herd. It turns out, everyone was trend trading because it was actually better. Or maybe it is "better" just because everyone is doing it (thus the psychological effect).

Mean reversion seems more complex, see #2 here: http://www.financial-hacker.com/build-b ... d-systems/

I even purchased a great range and still couldn't make it work.

Maybe if I revisited it now I would be able to incorporate seasonality and other new knowledge to make it profitable.

The saying of "cut your losses soon and let your winners ride" is more easily applied to trend trading than to mean reversion. It seems that you need those big winners in order to offset the more plentiful (but smaller) losing trades. With mean reversion, you don't have such opportunity as your profit targets are fixed.

For example, just scrolling up my posts in this thread, you can see I was wanting to bail on the GBPJPY trades when it was making $500, but "let my winners ride" and ended up with $900 instead.

I think the use of CSS or CCFp varies depending on strategies. I see continued reference to trends where my interest, for now, lay more with reversals or mean reversion.

[BostonEJ]

It seems that you need those big winners in order to offset the more plentiful (but smaller) losing trades. With mean reversion, you don't have such opportunity as your profit targets are fixed.

I like the higher frequency/win % of mean reversion and smaller profits vs. lower frequency bigger win $ of trend/momentum trading. Markets range more often then they trend so I have more chances of recovering in the case a trade goes against me. However, my ideal situation is two strategies where I trade mean and momentum. But mean is my first objective and trend will be behind it.

[afeudale]

I had the exact same sentiment....i.e. markets range more often than trend. I just couldn't make it work, no matter what I threw at it. Hopefully you have better luck.

I would say that pair selection is just as important with a mean reverting strategy as it is with trend strategies.

Have you chosen specific pairs? If so, how did you select them?

I like the higher frequency/win % of mean reversion and smaller profits vs. lower frequency bigger win $ of trend/momentum trading. Markets range more often then they trend so I have more chances of recovering in the case a trade goes against me. However, my ideal situation is two strategies where I trade mean and momentum. But mean is my first objective and trend will be behind it.

[BostonEJ]

I had the exact same sentiment....i.e. markets range more often than trend. I just couldn't make it work, no matter what I threw at it. Hopefully you have better luck.

I would say that pair selection is just as important with a mean reverting strategy as it is with trend strategies.

Have you chosen specific pairs? If so, how did you select them?

I don't think I'll be focusing on certain pairs as a rule but based on correlation over a period. Correlations tend to change. Pairs that correlate are likely to move similarly. Those would be my mean reversion pairs. Less correlated pairs would be the ones for trend/momentum trading.

[afeudale]

More trades from last night. GBPJPY and EURJPY. A EURUSD set of trades lost $70 and GBPUSD trades gained $5 after the positive USD news this morning that reversed a lot of the prevailing trends.

[SteveHopwood]

I see you are too thick to stop wittering on about craptesting, so here is the deal.

This thread is locked because wittering on about craptesting is informally banned at SHF. It will become formally banned if I ever work out how.

Witter on one more time about craptesting and I will remove your posting rights altogether. Or I might simply ban you; depends what sort of a mood I am in at the time. Tonight for instance, I am in a good one, so your posting rights and membership survive. Do not expect this favourable mood to continue. I can be in a bad one from time to time.

I know you are an idiot. I get that, but I trust that even somebody as dense as you will have got the message by the time you get to reading this sentence.

[SteveHopwood]

I see from the moderator's logs that afeudale attempted to post an 'Open letter to Steve' that was deleted by one of my mods. afeudale was then removed to the On Moderation group of SHF members unable to do anything more that read posts and download attachments. He/she is not banned but access is severely limited.

I did not see afeudale's letter; I do not care what it contained. His/her opinion is of no matter.

There is an obvious ethos here at SHF and members should go elsewhere if they do not agree with it. I refer to it further down.

Idiots like afeudale do occasionally crop up, with a forum death wish. They always achieve their wish. This dimwit was merely the latest in a long line.

I will not delete this thread. Instead I will leave it to remind later members that SHF is a meritocracy, not a democracy. Members earn the right to post freely by posting intelligently and helpfully in the first place. Stay quiet if you have nothing useful to say.

Remember my love of British Masterly Understatement when you read this next bit: I am moderately sure that I have mentioned this before.

Now we draw a line under this and return to intelligent discussion elsewhere.