tag:blogger.com,1999:blog-6758019036944570001.post4275253710732797031..comments2010-12-01T13:48:21.757-05:00Anunay Guptahttp://www.blogger.com/profile/03651129331296509296noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-6758019036944570001.post-2851792121436036582010-12-01T13:48:21.757-05:002010-12-01T13:48:21.757-05:00Very good journey and experience!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6758019036944570001.post-22738228842477091472009-11-29T23:52:11.811-05:002009-11-29T23:52:11.811-05:00Through my limited years of analytical exposure I've realized that one can solve around 90% of business problems using advanced data analytics (RFM, Value Migration, Trend Analysis) and intermediate statistical modeling (Regression - OLS & Logistic, Decision Tree, Cluster Analysis, Factor Analysis, Simple Time-Series Forecasting). The beauty of these methodologies is that the logic is easy to explain to the partners too and that to me is very important. We should strive to use approaches that don't appear as black boxes to our end-customers. Only when they understand the concept would they be more willing to implement the results coming out of these concepts into their decision making. This is when it becomes more of a "pull" rather than a "push".Pushonhttp://www.blogger.com/profile/10035602248709219812noreply@blogger.comtag:blogger.com,1999:blog-6758019036944570001.post-53277750759956744382009-11-29T11:27:43.185-05:002009-11-29T11:27:43.185-05:00I think business sense and outcome of real-time impact is always missed and mislead/mis-read sometimes by 'Statistical models' often considered in the analytics world as 'key skill' to work/survive - most of the times the real picture gets lost in the complex model(rather complexed) which fails reveal the bigger picture primarily due to lack of integration with business needs and blindly going by the technical/statistical expertisesaravananhttp://www.blogger.com/profile/02730774996817740349noreply@blogger.com