ETMA_The Evolution

By the way the English FA Cup will start in minutes, so the German, French Italian and Spanish cups.

Thanks for the info Bass..

Have you got a favourite team ?

I'm a Premiere League gunner fan .

Snap...

Roll on the European championships..

How to Build an Effective Trading System
(and Build Confidence that It Will Be Profitable)

#UCLfinal
http://www.telegraph.co.uk/football/...-atletico-mad/
Champions League Final from San Siro, Milano.
Enjoy,

Best.
B.R

Trend Following Wizards April
May 30th, 2016

Both the year and the month of April for the Wizards is mixed. No big trend (with no pun intended). Some are up, some are down April was generally negative while the year is overall positive (slightly only).

I watched the final Bass.. Madrid probably edged it in the end..

I've got some tickets to see England v Portugal ( Thursday eve ).. This is England's last game before the Euro's kick off so I'm looking forward to a big atmosphere
and hopefully a win...

then I cant wait for the tournament..

Yeah, England looks solid this Year, No losses in the qualifications, although i don't like Hodgson Management style, i think that England got a lot more talented coaches than Him Included but not limited to Redknapp, Pardew, Big Sam ( He did wonders saving The Black Cats from relegation) etc ...

Portfolio Risk Management .

UNNECESSARY LOSING QUARTERS (AND WHAT I HAVE LEARNED FROM THEM)

With broader diversification and better portfolio management, I am getting more and more close to avoiding from losing quarters. Yet, sometimes they can also come, and I must say that for me they have a positive value - I can always learn something new and motivate myself to work on further improvements and further testing of my ideas.

The unprofitable quarter itself is not a problem; it is something an experienced trader knows can come time to time - and it is necessary to reckon with it. Let me remind you that any trading approach in the world has its underperforming and unprofitable time periods - I have never heard of any strategy that wouldn’t have such periods (and from the nature of the markets, it is not even possible).

The sad moment comes when the trader realizes what he could have done in another, better way and quite often, with a little more advanced approach, could have completely avoided the losing quarter. What have I learned from some of mine, unnecessary losing quarters?

1) It is essential to control the overall profit of high-correlated strategies and also other strategies in the same group of markets

The biggest problem in one of my losing quarters was not the losses themselves, but the fact that in the losing day I had too many similar positions opened, from high-correlated strategies, and the loss for the day accumulated.

This experience led me to create something called GROUP HEAT. In short - to have a maximum risk amount set for all opened positions/strategies ingiven market group. By the market group, I mean strategies that are highly-correlated. For example, I can consider all my strategies for index markets to be such a group - because when there is a bad day for my breakout strategies in index markets, the loss is happening unnecessarily on too many fronts.

I was aware of the necessity of having GROUP HEAT already in the past, but, partially also from laziness, I have underestimated this step. Another reason was that when I was thinking about GROUP HEAT implementation, my index breakout strategies were doing well.

What is the definition of GROUP HEAT? It is a simple protection that checks that the total sum of stop losses of my currently opened positions in given market group is less than 3% of my current account (or any other value, depending on the size of your account and on your personal preferences). As soon as this limit for a group has been reached, no new positions will be opened.

This simple rule has helped me to save several unnecessary losing quarters or, at least, to reduce the losses dramatically.

2) I have learned that in the bad times I need to reduce the positions and in the good times to add up (to decrease the exposition when the strategy is in a drawdown)

From my previous tests and researches, I have found out that the best possible functional way in the bad times is to reduce my positions. Yes, the return from the drawdown is slower, but, what is more important, we protect our capital in the bad times and we don't let our whole portfolio go even into bigger drawdown.

Personally, I consider my strategies to be employees of my company: as long as they do a good job, they deserve more resources, a higher percentage of my trading account. When they do a lousy job, they deserve only a small percentage of my account.

So my trading approach in these situations is roughly following:

1. The deeper the system is in a drawdown, the lesser contracts you should allow it to trade.
2. As the system is getting back from the drawdown, you can slowly increase the number of contracts back to normal.
3. If you trade only 1 contract with each of your systems, consider temporary reduction of the number of systems.
4. I.e. systems with higher drawdown should have the lower priority over those that are performing well - and if there are bad times for the whole group (e.g. indexes), you should reduce the trading systems and the exposition in the given group, until the bad days for this group are over.
5. The last option is to increase or decrease the percentage of amount risked for every trade of a system. To do that you need to have a bigger account. The basic idea is again simple - the higher drawdown, the lower percentage of your account you allocate to the given strategy.

After lots of experiments, this philosophy has turned to work the best for me and has helped me many times in the weak quarters.

3) I have done a big mistake having too many long-biased strategies in index markets

The last reason for having losing quarters in many years of my life ATS trading was that I used to have in my portfolio much more long-biased index strategies then short-biased ones. Why? Simple because it is not so easy to develop a short-biased breakout strategy for index markets. The index markets tend to trend much more upwards than downwards so all potentially interesting short strategies have a small statistical sample of trades and that complicates its effective development and testing.

My experience from the past also showed that strategies with bias on one side are not a good solution. You need to build strategies trading both sides, or, at least, balance the long/short strategies in your portfolio - for every long-biased automated trading strategy you should find one that is short-biased.

If I have done that earlier, I could have saved some unnecessary losses.

I believe that when you take a moment to think about those 3 points described above, you will find out that they can help you to go through your bad times and improve your future trading results and stability of your profits.

Personally, I have paid quite a lot of unnecessary losses for these three lessons, but you can avoid them.

Author: Tom Nesnidal (more about me)

TRYING TO SMOOTH EQUITY WITH MULTIPLE SYSTEM FILTERS IS REALLY BAD IDEA
September 28, 2015

In the past, I used to be like most of the ATS traders and developers: I put an enormous amount of energy into each single strategy and had a vision of getting a perfect, smooth, stable equity. Yes, sometimes perfectionism is a good thing. Many times, however, it is a curse.

In the case of ATS development, it often does more harm than good. The more you try to make your strategy (and especially equity) perfect by adding more and more filters, the more you curvefit and overfit your strategy to the past data. The future live performance is, in most cases, fatal.

I could have probably saved plenty of money if I had known that with a basic portfolio of simple strategies (with minimum system filters) you could smooth your equity perfectly and at the same time keep your strategies really simple, with the minimal amount of system filters. It was just a few years ago when I realized about that and started spending much more time on perfecting the portfolio instead of my systems.
Now, let me give you some examples that show how a simple portfolio can really smooth your results and stability of the equity and returns hopefully, it will help you to see where the real effort needs to be placed on.

Lets start with the simplest example: a mini-diversification, just across different markets within one futures class in this case its going to be the indexes futures class.

I took a small variation of my system called BOSS ATS (you can download this system for free) and simply put together the version for emini S&P market, emini Dow Jones market, and emini NASDAQ market. So in this case, its simply one system diversified over 3 different markets within the same futures class - and surely with a pretty high correlation between the 3 systems.

Probably, the most important metric here is the Net Profit / DD ratio, which we want to have as high as possible. We simply want to have much higher profits than drawdowns. In the case of this min-portfolio, the ratio is 9.69. Not bad, but not great either. We can definitely do better, by also smoothing the equity.
Now, lets try a different example. Were still going to stick to 3 trading systems, but now, well diversify a little bit further. Well keep only the emini S&P (ES) version of BOSS ATS, and combine it with a completely different swing system designed by emini Russell (TF) market, plus with one system for Soybeans (S). So now there will be a diversification across 2 market futures classes (indexes ES and TF, plus grains S). The whole three systems are low correlated among each other. What happens now?

Not only is the equity much smoother all of sudden, without the need to touch the underlying systems, but most importantly, we have also increased the Net Profit / DD ratio significantly, which jumped to a very nice 18.46. This is way better! So, as you can see, you can just do a great job by making a very small portfolio, and only 3 systems can bring you a real improvement, without the risk of overfitting or curvefitting when trying to fix your systems by adding multiple filters. Finally, what happens when we combine much more systems together? Can it bring even better progress?

In the last example, I put a bunch of strategies together, both high correlated (like all the versions of BOSS) and really low correlated ones. To be more specific, I put together 7 systems over 3 futures classes (Indexes, Bonds, and Grains). So far, this has been the biggest jump you could have seen:

The equity looks really smooth now (and surely it can still get much better), without touching the underlying systems or increasing the danger of overfitting. Even better, the Net Profit / DD ratio is at 29.58, which is an absolutely excellent number, hardly achievable with a single ATS strategy that would be simple at the same time and contain minimum filters.

So, this is my lesson for today: Stop thinking in terms of a simple system and start thinking about the bigger picture. Dont repeat my mistakes.

Author: Tom Nesnidal (more about me)

Nice Bass. Multiple systems full steam ahead. :-)

we all like your personal library :nerd:

Hope it Helps,
Cheers.

Long Term Euro Charts Point to Trouble Ahead
May 24, 2016/by Peter Brandt

Long Term Euro Charts
For a considerable period of time I have held a very negative opinion on the Euro. The Euro currency has a complicated history. The Maastricht Treaty of 1992 obligated certain Euro nations to adopt a common currency – although it was not until Jan 2002 that currencies such as the D-Mark were officially replaced by the Euro currency mechanism. It is possible to create a proxy EUR that dates back to 1992. In fact, IMM and proxy price data exist back to the early 1970s. Several technical developments on the EUR graphs are worthy of note.

A 6-1/2 year descending triangle on the quarterly graph was completed by the decline in Jan 2015. This pattern has an unmet target of 84 cents, the area of the 2001 low. The monthly graph also displays a dominant multi- decade trend-line connecting back to the 1985 low. The decline in the first quarter of 2015 found support at this trend-line. It is interesting to note that the price range in the first quarter 2015 has contained the market ever since. The weekly graph displays a possible 27-month congestion. It is my contention that this congestion will eventually be resolved to the downside – and that the simultaneous violation of the lower boundary of this congestion zone and the multi-decade trend-line will tie into a crisis in the Euro currency mechanism.

The daily graph displays a completed 14-week rising wedge with a target of 1.0822.

Hedge Fund Style Research : The Felder Report

Webinar on Position Sizing and Trading System Evaluation .

Title: Money Management & System Evaluation Masters: How to Analyze a Trading System and Determine the Best Position Sizing Method to Boost Your Trading System Performance.

http://skilledacademy.com/courses/li...ctures/1124797

Andrea Unger - Skilled Academy
Known as the only Four-Time Trading World Champion (2008, 2009, 2010, and 2012), Andrea Unger is a full-time professional trader since 2001 and part of the Scientific Committee of SIAT (Italian Society of Technical Analysis).

Know Your System! Turning Data Mining from Bias to Benefit Through System Parameter Permutation.

Learning from the Insurance Industry: Using Stochastic Modeling to Improve Trading System Development

051 – Strategy Evaluation with Dave Walton.

Topics discussed

1. How the typical approaches to system development can introduce datamining bias without you knowing
2. The types of systems that can increase the chance of data mining bias and what to look for
3. How the method of splitting your out of sample data could be causing you to throw away good strategies
4. Out of sample, walk forward analysis and Monte Carlo do they actually reduce data mining bias?
5. The problems with using Monte Carlo analysis to assess strategy performance and why it doesnt protect from overfitting
6. System Parameter Permutation how to use it, why use the median, parameter range selection and new insights since the SPP paper was published
7. How System Parameter Randomization solves some of the issues of System Parameter Permutation
8. Stochastic modelling and how it can be used to determine if a rule is adding value to your strategy

We are on the same diet Bass.

Algorithmic Trading: The Play-at-Home Version
Building computer trading models has become the latest DIY craze

Building algorithmic trading models is becoming a new norm for DIY investors
Michael Singer - Monday, December 14, 2015