Mark Richards Male Chicago: Share

Blog Posts
Discussions (28)
Groups (26)
Photos
Photo Albums
Videos

Mark Richards's Apps

Mark Richards's Friends

View All

Mark Richards's Groups

Actuarial Analytics

34 members
New Books

47 members

Social Network Analytics

107 members
Decision Stats

98 members

Debit Card Fraud Analy...

7 members

View All

Mark Richards's Discussions

Mark Richards's Page

Latest Activity

Mark Richards replied to Manish's discussion 'Modeling Rare Events' in the group Analytical Techniques

"...a form of bootstrap sampling to boost your rare event cases higher for development purposes " Have you considered adaptive methods (e.g Gradient Boosting aka Salford System's "Tree Net")?

August 25

Mark Richards replied to Mikhail Gorelkin's discussion 'A new book "Computational Social Network Analysis. Trends, Tools and Research Advances" (November 2009)...' in the group Social Network Analytics

Do a search on "Introduction to Statistical Relational Learning" Getoor and Taskar. There's a ton of relevant hits. The book is avaialable on Amazon.

July 25

There's a table of contents here: http://www.cs.umd.edu/srl-book/

July 25

Mark Richards replied to Yi-Chun Tsai's discussion 'two-stage models to improve response prediction'

Hi Yi-Chun, Yes, it's pretty much that simple, and makes that much sense. I'd say Greene's "Econometrics" is a good place to start, and if you really want a deep dive then Maddala's book. You might just search on "Heckit" or "Heckman" and "Model"...

July 25

Mark Richards replied to Yi-Chun Tsai's discussion 'two-stage models to improve response prediction'

Yi-Chun's response is a fairly common approach (and a good description). The problem is that it assumes that the amount of the sale is independent of the probability of responding to the advertisement. If this assumption doesn't make sense then ...

July 22

I like Getoor and Taskar's recent book: http://mitpress.mit.edu/catalog/author/default.asp?aid=20060

July 22

Mark Richards and Navin singh are now friends

June 19

Mark Richards replied to Vincent Granville's discussion 'How to prevent scores from caking in scoring models?'

"I think there is a way to build Boosted Trees in R, but I'm not 100% sure." Grerg Ridgeway contributed "GBM" to cRan. But as I said above, if you only have a limited set of discrete covariate patterns, I don't think Boosting or bagging does any...

June 19

Profile Information

Short Bio:: 14 years statistician in various industries (pharma, credit, insurance, retail, cpg & entertainment)

My Website:: http://www.linkedin.com/pub/0/611/179

Years of Experience in Analytical Role:: 14

Comment Wall (14 comments)

You need to be a member of AnalyticBridge to add comments!

Join this social network

At 10:13am on February 15, 2009, Mike Kruger said…: Good to hear from you again. I hope the years have been kind to you.

Mike

At 6:43pm on September 13, 2008, fency said…: Hi Mark
How are you?long time no see?
Fency

At 3:58pm on September 8, 2008, Greg Schopp said…: Mark,

Tried calling you at ISO, you still there?

Greg

At 8:34am on June 26, 2008, Muhammad Riaz said…: hi Mark,
Thank you for inviting me to your group.
Best Wishes,
Muhammad

At 8:23am on June 23, 2008, Mary Parker said…: Thank you Mark. Let me know when I can be of assistance.

Mary

At 10:21am on June 5, 2008, Patrice Latinne said…: Thanks a lot, Mark for the great information you shared on my wall about emerging issues. I do appreciate. I'll provide some feedback as soon as possible.
Cheers,

Patrice

At 10:36pm on May 14, 2008, Vladimir Lukin said…: Mark,

Thanks for pointing that book by Soumen Chakrabarti "Mining the Web" good intro into Web and HTTP basics, and I am looking forward studying a little bit harder statistical techniques applied to web data.

VL

At 4:52pm on April 22, 2008, fency said…: first of all thanks for the invitation into ur friends group....ya am living in wheaton,IL...why did u ask me?