assignment 1

profileprojectek
3.docx

The number of goals per game scored by teams in a professional soccer league are shown below for a sample of 180 games:   

 

# of Goals

0

1

2

3

4

5

# of Games

57

53

41

19

8

2

 

We would like to conduct a chi-square goodness of fit test at the 5% level of significance to determine whether the number of goals scored per game by teams in the league has a Poisson distribution.  Since a Poisson variable can take any non-negative integer value, it is possible that the number of goals is even greater than 5 (even though there are no games in this sample for which X > 5).  As such, we must label the last column "http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%26ge%3B5".    

# of Goals

0

1

2

3

4

http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%26ge%3B5

# of Games

57 

53 

41 

19 

Expected Count

   

   

   

   

   

   

 

(a)  Since we don't know the value of the parameter http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%26lambda%3B, we must estimate it.  The estimated value of http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%26lambda%3B is  http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%40HAT%7B%26lambda%3B%7D%3D  . (b)  Enter the expected cell counts in the table above.  (When you are calculating probabilities for the Poisson distribution, round your probabilities to four decimal places.  Then round your expected cell counts to two decimal places.) Since  the last cell has an expected count less than 5, we must merge it with the cell to its left.  We will label this new cell "http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%26ge%3B4".  Our final merged table is shown below:  

# of Goals

0

1

2

3

http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%26ge%3B4 

# of Games

 57

53 

41 

19 

10 

Expected

   

   

   

   

   

Cell Chi-Square

   

   

   

   

   

(c)  Enter all expected counts and cell chi-square values (rounded to two decimal places). 

(d)  The value of the test statistic for the appropriate test of significance (rounded to two decimal places) is http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%26chi%3B%40SUP%7B2%7D%3D  . (e)  The P-value for the appropriate test of significance is between    and   . (f)  If we used the critical value method to conduct the test, the decision rule would be to reject H0 if  http://angel.bfwpub.com/intellipro/geteq.ashx?eqtext=%26chi%3B%40SUP%7B2%7D%26ge%3B  . (g)  What is the correct conclusion for this test?

 Fail to reject H0.  There is insufficient evidence that the number of goals per game by a team does not have a Poisson distribution.

 Reject H0.  There is insufficient evidence that the number of goals per game by a team does not have a Poisson distribution.

 Fail to reject H0.  There is sufficient evidence that the number of goals per game by a team does not have a Poisson distribution.

 Reject H0.  There is insufficient evidence that the number of goals per game by a team has a Poisson distribution.

 Fail to reject H0.  There is insufficient evidence that the number of goals per game by a team has a Poisson distribution.

 Reject H0.  There is sufficient evidence that the number of goals per game by a team has a Poisson distribution.

 Reject H0.  There is sufficient evidence that the number of goals per game by a team does not have a Poisson distribution.

 Fail to reject H0.  There is sufficient evidence that the number of goals per game by a team has a Poisson distribution.