thread: Synthetic hedges, cointegration, mean reversion and similar stuff

I think cointegration just does not work (anymore) on currencies. The job performed here others and by 7bit is a fantastic proof of concept and of great acedemic value and I'm sure can be applied everywhere.

I am not sure how cointegration work for a single set of markets and not another. I really don't think that's the matter and that I mean no disrespect. What is of difficulty is that cointegration works *some* of the time and not others. My experience is that are fantastic for range bound action although not trending markets. All things considered, all LM methods are based on the foundation that a mean is there. This can't hold true over longer intervals.

I read an excellent excuse of unit roots as if they were describing a blood flow of a river.
Just like the flows of a river, the directional bias of the individual flows of a basket of currencies change. At particular points in the river, the flows are centering around a loion. AT some points they're currently accelerating away from this stage. This is the foundation of the unit root test.

I have not researched ADF in depth to comprehend what its calling but based on my fundamental understanding I don't believe this accurately helps us discovers unit roots where the ebbs and flows from the river are many. I believe we have to visit the versions or Johanneson to become accurate unit roots.

Next Im a huge believer in HTF analysis. If we discover a basket is cointegrated on bigger (h1 ) timeframes, it opens up the potential for trading on the shorter timeframes when information from these timeframes drops out of synch.

None of this has been accounted for in anything I have read so far. Neither has been rebalancing, range bound checks, SR, etc..

Cointegration is a tool. It is not the holy grail. Managing the tool effectively should get us to where we want to be get which is to find an automated bot that eliminates the need to watch the markets constantly and still operate profitably.

FX-- I sent you an email. I am game.

quote I am unsure how cointegration work well for a single set of markets and not another. I don't think that's the issue and that I mean no disrespect. What is of difficulty is that cointegration works *some* of their time and not others. My experience is that are great for range bound activity but not trending markets. All LM approaches are based on the foundation that a mean that was central is there. This can't hold true over longer intervals. I read a fantastic excuse of unit roots as if they were describing a the flow of a.. .

I hear you. . Maybe I simply threw in the towel with co-int and fx, but the fx market has become so efficient over the past few years that such relatively easy techniques don't work anymore together with FPI /triangular arb etc..

If anybody is interested, http://arb-maker.com/archives/2677 pointed me to a recent paper called http://www.ljmu.ac.uk/Images_Everyon...DJLAS_0410.pdf that contains on pages 16 to 23 a few rather interesting test results and decisions. The abstract:

Cointegration is a technique that's been used for some time to optimise equity portfolios, but there is limited evidence of its appliion in managing currency portfolios. This study examines whether there's any value to be gained by using cointegration based egies to optimise currency portfolios which are U.S. dollar, Euro and Sterling based respectively. We construct #8216;major currency pair (MCP) monitoring #8217; portfolios into replie the timeless index monitoring egy commonly applied to equity portfolios, overcoming the absence of an 8216;index#8217; for currencies using the most frequently traded currency pair for each portfolio, specifically the EUR/USD for the U.S. dollar and Euro portfolios, whilst the GBP/USD is employed for the sterling portfolios. We compare the out-of-sample performances of those portfolios to easy benchmark techniques of optimisation. The results are encouraging, with the discovery of relationships that are long-run adding value for the portfolios that are sterling. Because of the generally low temperament of the cointegration portfoliosthey are leveraged to coordinate with the returns being offered to investors from the benchmarks, since they generally provided better risk-adjusted returns. They also improve portfolio stability.

There are many side issues which are also mentioned, like whether profits may nevertheless be derived from applying simple trading rules to currency trades, and if so what would be the top currencies to invest in and how much should an investor internationally diversify their portfolios. A few of the results found are reassuring, and purpose towards the value supplied by #8216;diminished #8217; currencies particularly the currencies, with the Swedish Krona appearing to boost the risk-adjusted returns of portfolios. Technical analysis is shown to still have a part to play in currency trading.

Is this system still profitable? No more update about this egy?

Exotics are very nicely co-integrated and stationary i.e USDSEK USDNOK, USDCZK USDHUF etc.. The spreads of them in the retail market are so big that it makes it impossible to make decent profits when seeking to arbitrage.

The rest of the pairs are not co-integrated like GBPUSD EURUSD etc.. These sort of pairs are not co-integrated at all.

I've run ADF tests on virtually all tradeable pairs and there is 8-10 pairs exotics which are co-integrated, however again, spread is so big on both that it is impossible to profit nicely.

Exotics are extremely well co-integrated and stationary i.e USDSEK USDNOK, USDCZK USDHUF etc.. No matter how the spreads of them in the retail market are so large that it makes it impossible to make profits when seeking to arbitrage. All other pairs aren't co-integrated like GBPUSD EURUSD etc.. These type of pairs aren't co-integrated in any way. ADF tests've run on almost all tradeable pairs also there's 8-10 pairs exotics that are co-integrated, however again, spread is so large on both that it's impossible to profit well.

Is there no broker provide low spread for exotic pairs? Therefore do you use this egy?

The variations on my website are the currrent versions and I use then myself currently. They contain no errors (besides maybe the fact that the whole approach of using a linear regression and hoping for useful results might be flawed). I'm still hoping that someone else (who is experienced in R and in numbers than me) will take the present code and replace the easy regression with something else or otherwise enhance the script and publish the modified version here. I'm doing some experiments but have nothing prepared for publishing...

Truly wonderful and really much appreciated. Love your job if it adds value to this thread and will post my progress.

One thing I intend to look at is utilizing TLS instead of OLS.

You've earned my vote voucher, though I can not issue a coupon.

Thanks again, Snow

One thing I intend to look at is using TLS instead of OLS.

TLS=absolute least square?
Which norm are you going to utilize to mesure the distance between two points?

picture here's another experiment just throwing each the majors together. I run a custom VSA template around the spot dollar index utilizing sigtrader. Seeing that it was in the center of its range this morning prior to the NY stock market open, I entered the recommended arbomat transactions while the price line was within 1 deviation of the blue core. Thus profitable.

From what I collect, I would have traded that the exact other way around.

library(zoo)
library(tseries)
backlt;-1000
nowlt;-1
cthis lt;- 1
x lt;- scan(c:/regdata.txt)
regressors lt;- matrix(x,ncol=6)
regressand lt;- regressors[, cthis]
regressors[, cthis] lt;- rep(0, back)
y lt;- regressand
x lt;- regressors
version lt;- lm(y ~ x 0)
summary(model)
stddev lt;- sd(resid(model)) coef(model)[-1]
//Plotting part
pred lt;- as.vector(predict(model, newdata=data.frame(x=I(regressors))))
curve lt;- rev(regressand-pred)
linea lt;- 0
lineb lt;- 0
plot(curve, kind ='l', ylab=”, xlab=”, main=”, sub=”, col='cornflowerblue')....

Hello, would you explain the outcomes of the script?
I see the plot, and I get this lineup

x2 x3 x4 x5 x6
0.580635548 -0.003192527 -0.001234799 0.995316248 -0.574710354

Does it means that I should Buy 1 lot of first pair, Buy 0.58 lots of second pair, Sell 0.003 lots of third, Sell 0.001 of fourth, Buy 0.99 of fifth and Economy 0.57 of the last pair?

Due