Thursday, October 11, 2007

Identifying Historical Trading Patterns and More Thursday Thoughts

* Patterns With an Edge - I am getting a fair number of requests for posts re: how to conduct simple backtesting of market patterns using Excel. My own sense is that this would be best accomplished through a hands-on, skills-building teaching session rather than a blog post. If this would be of interest to you, please leave a comment for this blog post and let's see what we can organize.

* Hot Sectors - Excellent research from Charles Kirk on how stock screening illuminates sector strength.

* Is The Worst Over for Banks? Several excellent perspectives are linked by Abnormal Returns.

* Finding an Edge Among Stocks That Have High Short Interest - Fascinating findings reported by the CXO Advisory blog.

* Partying Like It's 1997 - Bespoke Investment Group finds a link between today's market and that of 1997.

* Rising Inventory - Trader Mike updates his links with some interesting views, including a look at the rising inventory of unsold homes from Countrywide.

* Finding Yield - StockPickr identifies companies that have increased their dividends lately and who is buying those shares.

20 comments:

Raymond said...

Hi Brett, I would love that and would clear just about anything from my schedule to be able to attend !

RAndy

Simply Options Trader said...

Dr Brett,
I'll be interested in attending backtesting training session if you are able to organize one. Looking forward to it. Thanks!

gmu123 said...

Thanks Brett. I'll be interested in attending workshop on backtesting. Prefer live rather than webinar. Understand then it would be fees to pay for your effort and time.

Did you see current New Yorker magazine article:
http://www.newyorker.com/reporting/2007/10/15/071015fa_fact_cassidy

about Mr. Niederhoffer. In that article, he talks about market edge with historical patterns, etc. Amazing stuff. Similar to your way of looking at the market through a prism of data.

Thanks again.

craig said...

Dr. Brett

I would thoroghly enjoy such a opportunity. I will also take this time to say thank you for all your work here.

Sincerely,

Craig Garbie

Jeff said...

I would be most interested in, and grateful to any instruction on that subject you'd be willing to offer!

Srinivas said...

Brett,

I've been expecting this session for quite long time. If I remember it correctly, you wanted to do this as a webinar.

-Srinivas

BaiBu said...

Doc, will join your training class. Thanks, Mitsu.

The Piker said...

Dr. Brett:

Would you kindly reconcile what you are saying in this post with the statements that you made in the Wonderland post:

"Trendlines? Oscillators? Chart patterns? Wave formations? Angles? All of those figure prominently in the books, seminars, and magazine articles that dominate the mass market.

Yet I have yet to see a successful trader at any of these firms use these tools. You would think, by sheer chance, having worked with well over 100 traders personally and closely, I'd find *someone* who trades the way the books and magazines describe.

But no. I have found none."


IF not a single successful trader you met used such patterns, then why are you writing about how to identify historical patterns through backtesting?

Jeff said...

Count me in!

Johan said...

Sounds very interesting! Thanks!

Louminn said...

I would be very interested in a class on this topic. In the past, I have tried to make spreadsheets from the ideas you have generated on this blog. This would be extremely helpfull. I would prefer a webinar, unless your travelling to Minnesota!! Thanks for all of your help Brett.

Dion said...

I would be very interesting in learning about how to conduct backtests of market patterns using Excel.

Regards,

dp

elegy said...

I think that would be great. I'd attend.

James said...

I'd love to attend.

best to you brett,

james

st said...

count me in as well.

BirdMan said...

Brett
Please count me in!
Thanks.
Dave

Brett Steenbarger, Ph.D. said...

Dear Readers,

I just got back from CT working with traders and wanted to express my appreciation for the interest in a session devoted to backtesting. I'll definitely put something together; the big challenge is whether to conduct it as an online event or live. I generally prefer online for convenience sake and lower cost to traders, but live would allow for much greater interaction.

I'll follow up with plans shortly!

Brett

Brett Steenbarger, Ph.D. said...

Hi GMU123,

Thanks for the link. I learned a great deal from Victor Niederhoffer, not least of which is a respect for a scientific approach to trading.

Brett

Brett Steenbarger, Ph.D. said...

Hi Piker,

That's a great question, and I can see where it might be confusing. If you're familiar with statistics, I can tell you that I primarily test for interaction effects in the data rather than linear main effects. That is what makes the historical pattern analysis relational.

Brett

MidKnight said...

Outstanding Brett. I had often had the same thoughts that NQTrader posted previously.

Best regards,
MK