STAT 475: Data Analysis with SAS
Spring 08: Final exam


NOTE: Example format for your answer code.

	/*******************************
  		STAT 475 Final Exam
  		YOUR NAME (YOUR ID)
	********************************/

	***** PROBLEM #1 ***********;

	YOUR SAS CODE HERE;


	***** OUTPUT ***************;

	/*

	PUT YOUR OUTPUT HERE (copy and paste from output window)

	*/

	***** END OF PROBLEM #1 ****;


Repeat for each problem. Each problem should start on a new page.



1)  Suppose the following data (final08_1.txt)

Bob A Johnson  MATH STAT
 542
 345
 555
 544
 2 1
 33 
 455
Mike H Chen  APPLIED REGRESSION
 125
 345
 2 2
 222
 355
 442
  11
 232
 555
Mike H Chen  EXP Design Analysis
 554
 344
 445
 333
 1  
  22
 444
 555
Jane D Scheaffer  STAT INFERENCE
 455
 555
 54 
 344
 554
Jane D Scheaffer  BIOSTATISTICS
 233
 333
 211
 334
 442


which consists of teachers name, course title, followed by student rating on 3 questions.
For example there were 7 students in Bob's class.  The first student rated Bob 5 on the first
question, 4 on the second question and 2 on the third question. Note that there are some missing responses.
Also note that there are two spaces between name and course title.

Save the data and read into a SAS data set whose observations consist of teacher name, course name, and
the 3 ratings. Then,


  (a) For each teacher/course compute the average score for each question, and the overall average (of the 3 questions).

  (b) Find the teacher/course with the best OVERALL rating. 




2) Solve the equation  G = (G1, G2) = (0, 0) using PROC IML, where
                          

    G = (G1, G2) is as defined above:

                G1(x, y) = x - cos(y),

                G2(x, y) = y - exp(x).
   


3) Write a Macro function which calculates cumulative probability under Binomial distribution with probability function

   f(x) = [n!/(x!*(n-x)!)]* p**x * (1-p)**(n-x) and the cumulative probability is

   F(x) = f(0) + f(1) + ... + f(x-1) + f(x).

   Test your function with n=10, p=0.3. Your output should look like

   
	Cumulative Binomial Table with n = 10 and p = 0.3      
	
		x	F(x)
        	____________
 		0	....
		1	....
		2	....
		3	....
		4	....
		5	....
		6	....
		7	....
		8	....
		9	....
		10	....
	

  Use F5.3 format for the probabilities. Note that FACT(n) can be used for n!.