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
- /
- Interpreting results of statistical procedures

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
- Highlight
- Email to a Friend
- Report Inappropriate Content

08-06-2017 09:47 PM - edited 08-06-2017 09:49 PM

Hello

I am a sas programmer and some times do statsitical analysis.

I had studied statistics quite some times back

I intend to revise my fundamentals for inferential statistics particularly ttest, regression analysis and validating regression model, Anova. (I do have a personal copy of Statistics by example by Ron Cody but that in no way serves the purpose, though it is a very good and useful book.).

Can any body suggest a book where I can properly interpret the result.

For examplle if i am interpreting a proc ttest what does Pr > T stands for OR what shouldbe the appropriate value.

Not looking for indepth mathematics

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

Posted in reply to thesasuser

08-07-2017 02:35 AM

If you are looking for a good statistics brush-up book (and you already have Ron Cody's at hand), I recommend "Discovering Statistics Using SAS" by Andy Fields and Jeremy Miles.

Otherwise, SAS has a free statistics e-learning course, that would be my no1 choice

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

Posted in reply to thesasuser

08-07-2017 08:38 AM - edited 08-07-2017 08:49 AM

Assuming you have a hypothesis H0: a=0 .

Under the condition of hypothesis of H0 was right ,you can get a estimator (E0) for H0 .

P value stands for the probability of estimator (E) greater than E0 .

If P <0.05 mean there are very few estimator E is greater than E0, which means E0 is very big,therefore you should reject H0.

The criteria of P value is usually set at 0.05