# correlation coefficient

12 messages
Open this post in threaded view
|

## correlation coefficient

 Could someone please explain to me how to go about solving such questions. ISI 2007 24. For a given bivariate data set ) (xi , yi ; i =1, 2,..,n ), the squared correlation coefficient  ( r^2) between x^2  and y is found to be 1. Which of the following statements is the most appropriate? (A) In the (x, y) scatter diagram, all points lie on a straight line. (B) In the (x, y) scatter diagram, all points lie on the curve y = x^2 (C) In the (x, y) scatter diagram, all points lie on the curve y = a + bx^2 , a > 0, b > 0. (D) In the (x, y) scatter diagram, all points lie on the curve y = a + bx^2 , a, b any real numbers 15. The correlation coefficient r for the following five pairs of observations x 5 1 4 3  2 y 0 4 2 0 -1  satisfies   (A) r > 0; (B) r < −0.5; (C) −0.5 < r < 0; (D) r = 0. Please reply. thanks
Open this post in threaded view
|

## Re: correlation coefficient

 For ques no. 24 it is given that corr b/w y and x^2 is 1 mens that in a scatter diagram (x,y) the points lie on the curve y=x^2 but since corr coefficient is independent of change of origin and scale thus it shud b option d Akshay Jain Masters in Economics Delhi School of Economics 2013-15
Open this post in threaded view
|

## Re: correlation coefficient

 Yeah got it. What about the next question. The data looks independent so shouldn't r=0? How do we calculate r in that case?
Open this post in threaded view
|

## Re: correlation coefficient

 In reply to this post by Akshay Jain Hi Akshay,  the graph of y=x^2 is a parabola so they should not be correlated? Also the answer posted by amit sir is option c). Please reply
Open this post in threaded view
|

## Re: correlation coefficient

 In reply to this post by Arushi Here v have y=x^2 (since the coefficient of determination =1 b/w y and x^2) and if v calculate the corr b/w y and x^2 will comw out to be 1 (underoot of coefficient of determination) so if we plot x and y then it will comw out to be a parabola but m not getting vy sir has given c  option....i dont no where m makin the mistake coz correlation is independent of origin and scale so a,b can be any real numbers Akshay Jain Masters in Economics Delhi School of Economics 2013-15
Open this post in threaded view
|

## Re: correlation coefficient

 SIR pls correct me where m going wrong..... Akshay Jain Masters in Economics Delhi School of Economics 2013-15
Open this post in threaded view
|

## Re: correlation coefficient

 In reply to this post by Arushi i think its 15(B) “Operator! Give me the number for 911!”
Open this post in threaded view
|

## Re: correlation coefficient

 Y = x^2. so y cannot take negative values. hence a,b>0On Fri, May 9, 2014 at 6:12 PM, tsuki [via Discussion forum] wrote: i think its 15(B) If you reply to this email, your message will be added to the discussion below: http://discussion-forum.2150183.n2.nabble.com/correlation-coefficient-tp7588637p7589145.html To start a new topic under General Discussions, email [hidden email] To unsubscribe from Discussion forum, click here. NAML
Open this post in threaded view
|

## Re: correlation coefficient

 In reply to this post by Akshay Jain Haan..got it..Thanks
Open this post in threaded view
|

## Re: correlation coefficient

 In reply to this post by vandita24x7 Thanks Vandita