Results 1 to 3 of 3

Math Help - Linear regression

  1. #1
    Member
    Joined
    Nov 2006
    Posts
    139

    Linear regression

    Okay, this chapter is wearing me out. I am not getting anything of all this. Let's start from a basic thing.

    It says "compute SSr and SSE and (s_e)^2 and the coefficient of determination, given the following statisics computed form a random sample of pairs od X and Y observations (let's take just one)

    Sum from i=1 to n of (y_i - y)^2 = 100,000
    r^2 = 0.50
    n= 52

    No clue no clue..

    Is there anybody who can explain me the simple regression. I cannot see the point of all this. Please. Thank you.
    Last edited by 0123; December 30th 2007 at 10:49 AM.
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Grand Panjandrum
    Joined
    Nov 2005
    From
    someplace
    Posts
    14,972
    Thanks
    4
    Quote Originally Posted by 0123 View Post
    Okay, this chapter is wearing me out. I am not getting anything of all this. Let's start from a basic thing.

    It says "compute SSr and SSE and (s_e)^2 and the coefficient of determination, given the following statisics computed form a random sample of pairs od X and Y observations (let's take just one)

    Sum from i=1 to n of (y_i - y)^2 = 100,000
    r^2 = 0.50
    n= 52

    No clue no clue..

    Is there anybody who can explain me the simple regression. I cannot see the point of all this. Please. Thank you.
    I can't follow the notation here. You might want to look at some other
    reference on simple linear regression if you find text book difficult to follow.

    Look here.

    RonL
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Member
    Joined
    Nov 2006
    Posts
    139
    Okay thing are getting a very little clearer.

    Let us take again that excercice:
    "compute SSr and SSE and (s_e)^2 and the coefficient of determination, given the following statisics computed form a random sample of pairs od X and Y observations (let's take just one)

    Sum from i=1 to n of (y_i - y)^2 = 100,000
    r^2 = 0.50
    n= 52


    Okay. So. I understood more or less what it is asking. SSR sum square regression; SSE sum square errors. (S_e)^2 is = SSE/(n-2) the variance.

    Okay. (S_e)^2 I get with that formula. So I need to know SSE. SSE I get through this formula SSE= Sum of [y_i - (b_o + b_1*x_i) ]^2.
    And SSR= (b_1)^2* Sum of (x_i - x_bar)^2 (by definition)
    Okay SSR I can know subtracting SSE from 100,000.

    But this means I need to know b_1. But b_1= r*s_y/s_x ( b_1 is the slope coefficient estimator)

    R I have it. But s_y and s_x (the standard deviations) I don't. The text doe not give them. So how do I get them?

    Thanks for your help.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. perpendicular linear span (linear regression)
    Posted in the Advanced Algebra Forum
    Replies: 0
    Last Post: October 22nd 2011, 03:56 PM
  2. Linear Regression
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: August 25th 2010, 06:49 AM
  3. Regression fundamentals (linear, non linear, simple, multiple)
    Posted in the Advanced Statistics Forum
    Replies: 13
    Last Post: November 12th 2009, 06:47 AM
  4. Regression SS in multiple linear regression
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: June 20th 2009, 12:23 PM
  5. linear regression
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: May 16th 2008, 04:36 AM

Search Tags


/mathhelpforum @mathhelpforum