AnalyticBridge
Data Intelligence, Business Analytics
Hariharan Sunder's Groups
-
Indian Chapter
214 members
-
SAS Network
450 members
-
Analytical Techniques
381 members
-
Web Analytics
453 members
-
Data Mining
402 members
Hariharan Sunder's Discussions
Time dependent variables in logistic regression
1 Reply
Hi all,I need your thoughts regarding a logistic regression model i am building to predict customers propensity to churn.I have users of all tenure in the model sample ranging from 1 month onwards…Continue
Started this discussion. Last reply by Nathan Wang Feb 11.
Predicting CLTV
I am working on a project to forecast CLTV and would like to get suggestions regarding possible approach I could adopt. Below is a description of my project:Our client is an e-retailer specializing…Continue
Started Mar 7, 2012
Adjusting for Oversampled Data
6 Replies
Hi guys,I have some doubts regarding oversampling. I am trying to predict the probability of a person to unsubscribe from emails for an online retailer. I plan to run a Logistic regression model for…Continue
Started this discussion. Last reply by Jeff Oct 14, 2010.
Logistic Regression - Hosmer Lemeshow test
11 Replies
I'm confused about the Hosmer and Lemeshow Goodness-of-Fit test in SASfor logit models. It is giving me some problems. It is testingsignificant (p<0.001), which as I understand implies my logit…Continue
Started this discussion. Last reply by Nur Andi Mar 8, 2012.
Hariharan Sunder's Page
Latest Activity
Hariharan Sunder shared Theophano Mitsa's blog post on Facebook
Nathan Wang replied to Hariharan Sunder's discussion Time dependent variables in logistic regression
"Try survival analysis."
Hariharan Sunder posted a discussion
Time dependent variables in logistic regression
Hi all,I need your thoughts regarding a logistic regression model i am building to predict customers propensity to churn.I have users of all tenure in the model sample ranging from 1 month onwards based on their date of joining. Does it make sense to include variables for users such as Activity in last 30 days, Activity in 30-60 days, Activity in 60-90 days and soon till say Activity in 270-300 days.For users with say only 2 month tenure most of these time dependent variables will be null. They…See More
Hariharan Sunder commented on Vincent Granville's blog post The 8 worst predictive modeling techniques
"Hi Vincent, great post and i completely agree you point of view. In fact i have seen people trying to predict customers' next purchase $ value using just linear regression. Completely blows me how did they even manage to do that.
But as…"
Nur Andi replied to Hariharan Sunder's discussion Logistic Regression - Hosmer Lemeshow test
"I'm also confused about this test for binary logistic. In my study, significant H-L stat has been solved follow suggestion on this treads. But, another problem came!. I try two different method for variable selection : Forwards and Backward.…"
Hariharan Sunder's discussion was featured
Predicting CLTV
I am working on a project to forecast CLTV and would like to get suggestions regarding possible approach I could adopt. Below is a description of my project:Our client is an e-retailer specializing in Flash sales of designer apparel. A person has to register on their website to make a purchase or even to browse the sales taking place.We have defined 2 types of people:Member: A member is someone who registers on their site but has not made any purchase till now.Customer: A customer is defined as…See More
Hariharan Sunder posted a discussion
Predicting CLTV
I am working on a project to forecast CLTV and would like to get suggestions regarding possible approach I could adopt. Below is a description of my project:Our client is an e-retailer specializing in Flash sales of designer apparel. A person has to register on their website to make a purchase or even to browse the sales taking place.We have defined 2 types of people:Member: A member is someone who registers on their site but has not made any purchase till now.Customer: A customer is defined as…See More
Jane Hu replied to Hariharan Sunder's discussion Logistic Regression - Hosmer Lemeshow test
"I have the same issues with this stat. I think it is sample size issue.
I used VIF option under proc reg to make sure those variables entered into logistic model do not highly correlated. So correlation is not an…"
Hariharan Sunder commented on Ayush Biyani's blog post low accuracy in out of time dataset
"Ayush,
Does your modeling sample consist of data collected from only one period in time? If so the model may be necessarily hold good on a out of time sample. I think your modeling sample needs to be much more random so as to include effects of…"
Hariharan Sunder replied to Hariharan Sunder's discussion Predicting Customer Purchase Behavior in the group Analytical Techniques
"Lokesh,
My problem is not to identify whether a member will become a customer or not. Regardless of whether a Member is going to make a purchase or not I have to classify them into one of the 4 divisions. So in order to classify the Members, I need…"
Lokesh Manohar replied to Hariharan Sunder's discussion Predicting Customer Purchase Behavior in the group Analytical Techniques
"First use logit to predict that a member will become a customer.
Then use multinomial logit to predict which product the customer will buy.
In the first step, besides other variables the characteristics of the customer while they were still a member…"
Hariharan Sunder added a discussion to the group Analytical Techniques
Predicting Customer Purchase Behavior
Hi all, I am trying to build a model to predict purchase behavior. Below is a detailed description of my problem:For a online retailer we are trying to predict which division a prospective customer is most likely to buy. Currently they have 4 divisions: Men,Women,Home and Kids.Customer: A customer is defined as a user who has made atleast 1 purchase. Members : Member is user who is on file but has never made a purchase.So we want to build a model on current customer to predict which of the 4…See More
Jeff replied to Hariharan Sunder's discussion Adjusting for Oversampled Data
"Hari,
Are you building a predictive model? If so are you using the model to rank only or do you need *accurate* probabilities as well? If its the former, I would say there is no reason to be concerned with weighting the data back (validate the…"
Tom Wolfer replied to Hariharan Sunder's discussion Adjusting for Oversampled Data
"Hari,
1. What is your Chi-Square statistic (critical value)?
2. What is your assumption regarding Degrees of Freedom for your crosstab test?
3. Are you conducting a one or two-sided test?
Thanks,
Tom"
Hariharan Sunder replied to Hariharan Sunder's discussion Adjusting for Oversampled Data
"Hi Guys,
One more doubt regarding Cross-tabs. When I run a 2x2 cross-tab, I get a P-Value<0.001 but the Phi-coefficient is very very low at 0.0063. When i just look at p-value it might show that both variables are correlated but my…"
Tom Wolfer replied to Hariharan Sunder's discussion Adjusting for Oversampled Data
"Hari,
Yes, that is exactly what I would do: applying the .4 and 1.2 correction factors, respectively, to the 25% (unsubscribers) and 75% (subscribers) sample representations will allow you to then conduct crosstabs and other analysis as if you were…"
Profile Information
- Short Bio:
- I am working as a Senior Business Analyst in Mu Sigma Business Solutions. Currently working on projects involving Churn prediction, Cross-sell/Up-sell models, Email Personalization for a leading e-retailer.
• Master of Science - Industrial Engineering & Management (August 2008)
Specialization: Supply Chain Management
Minor: Data Mining (OSU/SAS Data Mining Certificate)
Oklahoma State University, Stillwater, USA
• Bachelor of Engineering – Mechanical Production Engineering (June 2006)
Osmania University, India
• Certifications: Certified Predictive Modeler using SAS Enterprise Miner 5.2, by the SAS Institute Inc
• Specialities: Base SAS, SAS Enterprise Miner, SAS Enterprise Guide, SPSS, Minitab, AutoCAD, Solid Works, ARENA, Xpress Mosel,MS Project, SQL
- Field of Expertise:
- Business Analytics, Predictive Modeling, Data Mining, Operations Research, Statistical Programming
- Years of Experience in Analytical Role:
- 3
- Professional Status:
- Other
- Interests:
- Finding a New Position, Networking
- What is your Favorite Data Mining or Analytical Website?
- http://www.ats.ucla.edu/stat/sas/
- What Other Analytical Website do you Recommend?
- http://home.ubalt.edu/ntsbarsh/stat-data/Forecast.htm#radSatatistEr...
- Your Company:
- Mu Sigma
- Industry:
- Analytics
- How did you find out about AnalyticBridge?
Comment Wall
- No comments yet!
Top Content
Follow us
© 2013 AnalyticBridge.com is a subsidiary and dedicated channel of Data Science Central LLC
