|
PureBytes Links
Trading Reference Links
|
I got many requests asking for books on martingales. Therefore, I will
respond one last time to the entire list about martingales for those
interested.
John Sweeney's book titled "Maximum Adverse Excursion" talks about
martingales in a couple of the chapters and gives some sample spreadsheets.
I talked to John about martingales. His discussion about martingales in the
book is rather limited compared to his other "excellent" concepts in the book
but it is about the best thing that I have seen. He asked me to keep in
touch because he was interested in the work I was doing. Most research in
this area by mathematicians, that I have seen, is limited to the restrictive
typical gambling martingale. If I remember correctly, John only gives a
couple of ways the martingale can be constructed but try to get the concept
for how flexible your martingales can be when you read this book. If you
think hard enough and are a mathematical wizard, you will probably come up
with a hundred ideas that are better. He sounded somewhat biased against
martingales in his book but seemed to get very excited when I talked to him
about some things I had done. I think I purchased my book from Trader's
Press but you can probably buy it from Amazon.com
I had intended to hire a programmer to code software that would automatically
write 1000s of different martingale algorithms and test each one over
hundreds of stocks over 10+ years each. I realized this was a monumental
task because I have hired many programmers in my past businesses. I just
never seemed to have the time because of other life commitments. However, I
hacked away at some Excel spreadsheets about 3-4 years ago (with some help
from a professional VB programmer) and got very excited about the potential
when you do not limit your thinking to the typical gambling martingale
concepts in all the books that I have read. I have successfully traded with
some ideas that this limited programming created. However, I have been very
frustrated because of the potential I have not been able to tap because of
the absence of the code for historical testing. I never invest significant
capital on trading concepts unless I have done adequate backtesting. I want
code that will create over 100,000 trades per martingale algorithm on a
basket of 500 stocks, futures, etc. I am sure most of you would say that
this is overkill but that is what I have developed for backtesting "same bet
size" systems. I have done this with a "C" program linked to TradeStation
but I can't develop the code that automatically tests 1000s of martingale
algorithms with TS. One idea would be to use TS to create an entry and exit
system and then export the entry and exit bar dates/time to this program to
use with each martingale algorithm. Any ideas on this would be welcomed.
In a few weeks when my personal obligations decrease I would like to have a
discussion with the mathematical wizards, VB programmers (or C), and traders
that have used martingales. At that time I will share the work that I have
done with martingales. Perhaps, together, we can create a moneymaker. Those
interested that have the above experience, please respond privately and I
will get back in a couple of weeks when my personal obligations lighten.
Russ
|