There are some powerful tools available in WL (raw profit mode, MAE/MFE, position sizing etc). After using the product for a couple of years (great product keep up the good work guys), I do not believe I'm approaching system development properly and wonder how many other individuals fall into this category. My approach is I usually jump right in with entries, exits, stop losses, and position sizing and start varying parms.

There are books out there and I'm not opposed to reading some (I have one) but most seem to be generic. My question is more;

"What is the logical step by step approach people use to develop, test, measure the success of and implement a robust strategy in WL, specific to WL tools ? Use Raw Profit mode first ? Test entries separate from initial stop losses ? Use MAE/MFE to determine exits ? etc etc.. Then test out-of sample data.. then walk-forward testing ?

The above is in regard to testing a portfolio of stocks (not daytrading one stock each day), although the approach would seem to be fairly similar.

Any ideas, books or links would be appreciated.

thx

Hi Joe,

I have been working on just such a book for a while now. I teach Masters/PhD level Computational Finance at Bond University in Australia, and my students use WL (version 4) for their development. We will probably upgrade to version 5 next time I run the course (probably later this year), assuming there is a MC-Lab and NL for WL by then.

I started writing the book just before WL transmogrified from version 4 to version 5, hence the focus on v4. I don't think you will have much trouble carrying out the same functions in WL5. Basically, the main tools needed are:

1. WL,
2. Access to a statistical package (such as SPSS, etc)
3. NL (my students also develop neural strategies, which are covered - its not necessary, but if you want to follow this part of the book, you need NL).
4. MC-Lab

Bear in mind that the focus of the course is on "how" to develop successful systems (and is reasonably academic - purpose is as a textbook for students who want to do PhD work in Comp Finance, or are aiming to work for/with "trading" hedge funds), and it also covers details on fundamental (analysis) variables rather heavily.... as this is a particularly fruitful area which is overlooked by most tech quants. The book is also quite heavy with bibliography references, which you would need to chase up and read etc, to get ideas as to which variables you might want to base your systems on. Also bear in mind that the book is used to supplement 4 hour lectures and 2 hours workshops each week for 12 weeks, so there will be plenty of stuff not covered in the book!

If, after reading all the above caveats (!), you are interested in 'beta-trialing' my book, let me know. I can confirm that the majority of my students were (eventually!) able to develop successful strategies using the techniques described. Unfortunately, I can't replicate what we did in workshop time, but each workshop was focused on a chapter of the book. I can provide some degree of limnited support by email. Finally, your WL profile doesn't say where you are located, but I assume its not in Australia, so you should also know that all of the examples given are based on the Aus equity markets, and you would need to rework them for your own markets.

Let me know if you are interested.

Regards,
BruceV

thx Bruce. I might be interested in taking a look. I believe I have your email address and will send you a follow up.

I'm surprised with over 70 views no one else has commented. Interesting.

Bruce

If you can stomach another reader I would also be most interested to
test-drive your book
email :

CODE:
Please log in to see this code.

Bruce,
if you need more reviewers for your book feel free to email me (email is in my profile). i could comment on it in 2-3 weeks. i am a day trader (1.5 years) with very limited system development exposure.

Bruce,
re: your book review.
email it here.
garron@hotmail.com

ps: where are the traders?

Maybe because many of us don't like to advertise our email address on web boards?
At least that is my excuse.

Hi,

I have had a lot of responses "off-forum", and have a group trialing it already. Probably best to let them work through all my mistakes, then we can try for a second iteration. Thank you all for your interest.

Cheers,
Bruce

Guys, instead of revealing your e-mail addresses to the world (remember, it's kept hidden in your profile), why not use a tool like Whspr (there are several of a kind)?

I'm very intereste in brucev's course,too.
And I'm long in for your second iteration of the book.
below is my email: lindashan2000#yahoo.com (please repalce # with @)

Thanks in advance.

Hi Bruce, I'd also like to be a "beta" tester. bcsigal@gmail.com

Thanks

Burt

I have been one of the beta-tester for BruceV’s new book: “Designing StockMarket Trading Systems”. His book has for intended audience future hedge fund managers and as such it had to cover all of the subjects needed and necessary to design and executed automated profitable trading systems with emphasis on profitable. Wealth-Lab being the strategy testing environment of choice.

In his book, Bruce, does cover all aspects of the game. From initial setup to designing sophisticated trading systems. Everything is considered, from fundamental to technical variables with for only objective to design strategies that can be automated as much as possible.

I currently don’t know how he will make his book available, I do have some ideas, but I’ll wait for the release. What I know is that such a book would have saved me a lot time some years back as it is a lot more expensive to learn by doing real time than by following in the footsteps of other knowledgeable traders. For newbie to just below “guru” status, I would not hesitate to recommend Bruce’s book as it can not only save you money, it can save you something more precious: your time.

Happy trading.

Hi Bruce,

Has the book been released yet? If not, anyway to get a copy by email?

Thanks,

Mike

Hi Bruce,

i am very interesting in your book , please include me in your list.
uriel-at-tedgi.com

Regards
Uriel

Hi Mike and Uriel,

there appears to be little interest by publishers in producing books about the stockmarket at the moment!

I am currently working to push the content into an e-book. I also run a boutique hedge fund (www.portercapital.com.au), which will soon also sell trading system signals (like a collective 2 for my hedge fund). I will most likely make the book available through that channel. More on this soon.

One thing of note though. I have had several people write to me over the last few weeks asking for examples. I work a lot with neural networks and monte-carlo simulations (which are also covered in the book). The Neuro-Lab tool (and the MC-Lab tool) don't work with WL5, so I still work in WL4.

Most of my book does not contain code - it contains research, ideas, and (most importantly) the methodology that I use to create trading systems. To me, coding is what you do at the end, not at the beginning.

I believe that systems design starts with ideas and research, not data mining and endless testing of entries and exits. I feel that all traders go through a process of trying out endless tweaks and tests of a vast variety of systems - constantly changing entries, exits, and money management settings. This type of approach must be replaced by research and logic, and implemented through a workable methodology. This is what I am trying to get accross in my book.

I also know that although this sounds sensible, it is not all that attractive to many 'traders'. I had 5 people 'trial' the book - I sent all 5 of them the first 3 chapters. Only 1 of them asked for the remaining chapters and actually worked through the methodology. That was Roland, and he has put his comments on the book and methodology into this post above.

I was not suprised that 4 out of the 5 did not pursue the methodology. It takes hard work to develop a profitable, good quality system. The idea that you can just grab a bit of code, and manipulate a few simple lines, and then create money seems to be endemic amongst would-be traders. These people seem to believe that they can spend a few hundred bucks and a couple of hours, and then beat professionals at their own game.

I do not believe it is possible to create successful systems without research, logic and methodology. Although I do not want to dissuade people from buying my book, I feel obliged to point out that if hard work bothers you, then you should spend your money on something else.

Happy to discuss further.

Cheers, and good trading,
Bruce

Hi Bruce,
I have read many books on building mechanical trading systems. From 'Breaking the Black Box" by Martin Pring (too basic to be of any value) to Robert Pardo's classic "Design, Testing and Optimization of Trading Stsrems" (a must read for anyone serious about building trading systems)

I completely agree with your approach of using research, logic and methodology rather than searching for random variables. I wish you every success in publishing your book and would love to read it and give you my feedback. How can I get my hands on a copy?

Cheers
Pishu

Bruce,

You should consider self-publishing or on-demand publishing, where your book is printed only when someone wants to purchase it. Then again you probably can't beat the cost of making your own PDF copy of the book.

Best regards,
Dave

Hi,

quick update: I now have a publisher for this work, so hopefully it should be available before too long! The publisher is http://www.harriman-house.com/pages/home.htm

I am now just working on 2 complete systems to include as fully disclosed examples in the book. One is based on fundamentals and one on technicals, and both will contain ANNs, so my objective is to show how to use the methodology to design a working system, using either fundamental or technical variables (or both!), and then to enhance the system using ANNs. The book also shows the correct way to benchmark, test etc from a trading/stats point of view, as that is something I have received quite a lot of emails about from WL users over the years.

Many thanks to those who trialed the book and methodology for me. The methodology has worked for me, it has worked for my students, and over the years it has worked for those WL users who have put in the effort and persevered. You can be sure that I am taking into account the feedback I received. Finally, in line with the comments I received, and a little pressure from the publisher, the book will no longer be "too" academic!

Thanks again,
Bruce