turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Regression technique if the sample size is very le...

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

12-14-2016 08:34 AM

Suppose if I have a sample size which is less than 200. What is the best regression technique I can use .. for linear regression ..

KS Statistics with ten decile

Poison?

linear regression?

Sameer

Accepted Solutions

Solution

12-16-2016
12:57 AM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to sameer112217

12-15-2016 10:08 PM

All Replies

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to sameer112217

12-14-2016 08:51 AM

The technique is more dependent on question and data rather than sample size.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to sameer112217

12-15-2016 03:17 AM

What kind of regression model you are using ? simple linear regression or Logistic regression ?

You could try Robust Regression

OR

Exact method:

proc genmod

exact ........

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Ksharp

12-15-2016 08:52 AM

logistic regression .. what does genmod does actually?

Solution

12-16-2016
12:57 AM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to sameer112217

12-15-2016 10:08 PM