Hypothesis testing

  1. 1. The problem statement, all variables and given/known data
    I have used OLS and found that:
    b0 = 6,85
    b1 = 3,88 with se= 0,1121
    n = 20


    Person x claims that b1 = 5
    Choose an alternative hypothesis. Does your estimated relationship support this claim?
    Use a 5 % significance level



    3. The attempt at a solution

    H0 = b1 = 5
    H1 b1!= 5
    t((1-a)/2,18) = 2,101

    t = ([tex]\overline{x}[/tex] - h0)/se(b1) = (3,88 - 5)/(0.1122) = -9.99


    t lies not in the interval +- 2,101 so I reject H0

    Is this correct?
     
  2. jcsd

  3. se==? and does "b0 = 6,85" means "b0=6.85"?
     
  4. Thanks for replying

    Se = standard error

    I'm not sure about what you mean with the last question, but b0 was point-estimated to be 6,85 or 6.85 if the comma was what you asked about.
     
  5. From what I know of Hypothesis test and student T distribution (from Statistics course which I am still undertaking), your answer is correct ie "t" lies not in the acceptance region but in the left rejection region thus "H0" is discarded and "Ha" is accepted instead.

    A very small error that i have noticed is that you were supposed to look for the range of acceptance region corresponding to "significance level =5% " and "degrees of freedom= n-1 =19" where as you seem to have looked it up for degrees of freedom = 20 which is not correct and might result in loss of a few marks in exams even though your answer is still correct.

    So
    t((1-a)/2,18) = 2.09 ..................(At least in the table i posses)

    I Hope you are satisfied with my reply.
     
  6. Thanks:)
     
Know someone interested in this topic? Share this thead via email, Google+, Twitter, or Facebook

Have something to add?