Which characteristic of a data set makes a linear regression model unreasonable?

a. A slope close to 0
b. A correlation coefficient close to 0
c. A slope close to -1
d. A correlation coefficient close to -1

Question

Which characteristic of a data set makes a linear regression model unreasonable?

a. A slope close to 0
b. A correlation coefficient close to 0
c. A slope close to -1
d. A correlation coefficient close to -1

jim_thompson5910 · Answer

Do you remember what the correlation coefficient means?

anonymous · Answer

It measures the strength and direction of a linear relationship between two variables, right?

jim_thompson5910 · Answer

exactly

jim_thompson5910 · Answer

so if r is close to +1, then what does that mean?

anonymous · Answer

Then that means it's a positive correlation?

jim_thompson5910 · Answer

a very strong one, yes

jim_thompson5910 · Answer

if it's close to -1, then you have a strong negative linear correlation

jim_thompson5910 · Answer

and r being close to 0 means that it's very likely there is no linear correlation (so maybe some other regression, like quadratic regression, would be used)

jim_thompson5910 · Answer

this page explains the concept and more http://mathbits.com/MathBits/TISection/Statistics2/correlation.htm

anonymous · Answer

Alright, I got b!

jim_thompson5910 · Answer

yep

anonymous · Answer

Thank you very much!

jim_thompson5910 · Answer

you're welcome