Subscribe to Vincent Granville's Weekly Digest:

Information

Decision Stats

Website: http://www.decisionstats.com
Members: 116
Latest Activity: Sep 3, 2012

Decision Stats Subscribe and Search Widgets


The map has been moved to www.decisionstats.com . In addition a customized search engine for 776 sites is here

http://www.google.com/coop/cse?cx=010190610432068918325:583oni6eute

With credit to Dr Vincent for motivating me to do this.




Discussion Forum

US Training Courses for Open Source Data Mining Software RapidMiner (October 2008) 2 Replies

Started by Ralf Klinkenberg. Last reply by Vincent Granville Sep 23, 2008.

http://feeds.feedburner.com/DecisionStats

Loading… Loading feed

Comment Wall

Comment

You need to be a member of Decision Stats to add comments!

Comment by Marybeth Haas on January 29, 2009 at 9:07am
As per Ajay's suggestion, I'm alerting the group to some new predictive modeling positions that my client has open. The client is on the East Coast and is a very secure and stable (no bad loans here!) personal and commercial lines insurer. The client has spent the last year or so concentrating on building the quantitative skill level of their team in Research and Development, and have buy in from the C-level to continue to do so.

Director of Online Marketing Analytics - This position will work on the front end strategy and back end campaign measurement of the direct to consumer channel. 3 years web analytics and predictive modeling experience.

VP of Research and Modeling - They prefer a quantitative PhD with insurance experience. This is the real heart of the business - the R&D group for personal lines. Manage a team of up to 7 in exploratory research.

Please get in touch with me for more details.

Thanks
Marybeth
marybeth@haasandriley.com
312-265-6280
Comment by Edith Ohri on September 18, 2008 at 6:09pm
Regarding "Where were their quants"
As a data mining person, I maintain that statistics and applied math are no good for analytics:
(a) they "look backward", thus if a fraud or an error has been paying off, such bad practices would be endorsed and recommended for next action!
(b) The quants models rely on prior assumptions, therefore are subjective and prone to human mistakes and manipulation. As such, they are useless for the purpose of early detection and for that matter - for detection of any new phenomena.
(c) The current models belong to the times before IT. They don't address complexity, and become so complicated that no body can run them effectively (not mentioning controlling them).

As I see it, we venture into the IT era, equipped with management tools that belong to a century ago. In these circumstances, in my opinion, the only rational thing to do seems to be, limiting the level of complexity of financial tools that should be permitted.

- Edith

http://users.actcom.co.il/~edit/#GT
Comment by Deepak Bansal on August 13, 2008 at 2:14am
Hi this is Deepak Bansal, BSc (H.) Stats, MA Bus. Eco. and completing MBA in Fin. from IBS Hyderabad (finishing in Feb 09). Looking for career opportunities in Investment research, financial analytics and data modeling etc.
 

Members (115)

 
 
 

Follow us

© 2013   AnalyticBridge.com is a subsidiary and dedicated channel of Data Science Central LLC

Badges  |  Report an Issue  |  Terms of Service