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
- /
- probability and Correlation can we combine them

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

11-06-2015 08:33 AM

Hi All,

Hope you are doing great.

I have a question i dont know if this is a right question to ask.

Is there a way to combine Correlation and Probability . My idea is that if something is correlated to something then what is the probability that will be exactly that much correlated.

For Example Y is depended Variable and X is Dependent .

Correlation = .70

What would be the probability that it will be .70 all the time ..is it a silly thought ?

Regards

Kanchan

Accepted Solutions

Solution

11-06-2015
09:23 AM

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

11-06-2015 08:44 AM - edited 11-06-2015 08:45 AM

This post might help: **Understanding your data- the difference between correlation and conditional probability**

http://versionone.vc/correlation-probability/

All Replies

Solution

11-06-2015
09:23 AM

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

11-06-2015 08:44 AM - edited 11-06-2015 08:45 AM

This post might help: **Understanding your data- the difference between correlation and conditional probability**

http://versionone.vc/correlation-probability/

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

11-06-2015 09:23 AM

Thanks this helped

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

11-06-2015 10:54 AM

I think you are asking about a parameter in the population and how it is "realized" in a sample taken from the population.

If you have two variables X and Y that are correlated with R=0.7, the question to ask is "if I draw a sample of N pairs of (x,y), what will the correlation be in this sample?" The field of statistics is essentially dedicated to this question and its generalizations. The basic answer is that if N is very large, then the sample correlation will be very close to 0.7. Statistics uses tools from mathematics to answer "how close." In general, the answer doesn't just depend on N, but also on the joint distribution of X and Y.