Ask Experts Questions for FREE Help!
  Advanced
Register  |  Log in  
   Ask    
 Answer  
  Help  

Ask QuestionsprogressAnswer QuestionsprogressBuild ReputationprogressBecome an Expert
 
Free Answers in 3 Easy Steps

Register Now
3 Steps

At Ask Me Help Desk you can ask questions in any topic and have them answered for free by our experts. To ask questions or participate in answering them you must register for a free account. By registering you will be able to:
  • Get free answers from experts in any of our 300+ topics.
  • Accept money for answers that you provide.
  • Communicate privately with other members (PM).
  • See fewer ads.

Home > Business & Careers > Accounting   »   linear regressions

 
Thread Tools Display Modes
Question
 
 
#1  
Old May 11, 2007, 02:10 PM
femaleinjax
New Member
femaleinjax is offline
 
Join Date: May 2007
Posts: 2
femaleinjax See this member's comment history on his/her Profile page.
linear regressions

Green Mountain College is a 5,000 student state-supported, four-year institution located in the mid-South. Physical facilities can accommodate another 1,000 students, and the college administration is attempting to estimate the added yearly cost of educating the additional students. The Business Manager of Green Mountain College has asked you to evaluate two linear regressions given below, and recommend the better one to her.



Where: SC = Student cost
CH = Cost per credit hour
IS = Incremental cost per student

Required:
a. Defend your choice of cost function (regression 1 or regression 2) for predicting added student educational costs per year.

Y = a + bx + e

Y = the amount of the dependent variable
A = a fixed quantity
X = the value for the independent variable
B = the unit variable cost
E = the estimation error

MY WORK BELOW:
REGRESSION 1 REGRESSION 2
R-squared = .594 R-squared = .707
t-value = t-value =
standard error = 117 standard error = 133
Y = a + bx + e Y = a + bx + e
Y = a + .594 x + 117 Y = a + .707 x + 133

Not sure how to get t-value or the rest of the formula.


b. What information value does the standard error of estimate (SE) have in this situation?
For situation 1, since the regression estimate is 4,303 and the SE is 117, there is reasonable confidence that the unknown actual value lies in the range 4303 +/- 117, between 4420 and 4186.
For situation 2, since the regression estimate is 3,800 and the SE is 133, there is reasonable confidence that the unknown actual value lies in the range 3,800 +/- 133, between 3933 and 3667.


Does my work look right?

Reply With Quote
 
     

Answers
 
 
Old May 11, 2007, 07:21 PM   #2  
Junior Member
gogosean is offline
 
Join Date: May 2007
Posts: 47
gogosean See this member's comment history on his/her Profile page.
Quote:
Originally Posted by femaleinjax
Green Mountain College is a 5,000 student state-supported, four-year institution located in the mid-South. Physical facilities can accommodate another 1,000 students, and the college administration is attempting to estimate the added yearly cost of educating the additional students. The Business Manager of Green Mountain College has asked you to evaluate two linear regressions given below, and recommend the better one to her.



Where: SC = Student cost
CH = Cost per credit hour
IS = Incremental cost per student

Required:
a. Defend your choice of cost function (regression 1 or regression 2) for predicting added student educational costs per year.

Y = a + bx + e

Y = the amount of the dependent variable
A = a fixed quantity
X = the value for the independent variable
B = the unit variable cost
E = the estimation error

MY WORK BELOW:
REGRESSION 1 REGRESSION 2
R-squared = .594 R-squared = .707
t-value = t-value =
standard error = 117 standard error = 133
Y = a + bx + e Y = a + bx + e
Y = a + .594 x + 117 Y = a + .707 x + 133

Not sure how to get t-value or the rest of the formula.


b. What information value does the standard error of estimate (SE) have in this situation?
For situation 1, since the regression estimate is 4,303 and the SE is 117, there is reasonable confidence that the unknown actual value lies in the range 4303 +/- 117, between 4420 and 4186.
For situation 2, since the regression estimate is 3,800 and the SE is 133, there is reasonable confidence that the unknown actual value lies in the range 3,800 +/- 133, between 3933 and 3667.


Does my work look right?
Here is one they don't tell you a lot. The Y intercept, which is A, is sometimes considered Fixed Cost. The R squared evaluates % variance that the IV accounts for the DV, so model 2 is a better fit. I would also include the interpretation of the (coefficients/beta weights). Also, you have t values for all predictors, what is <.10 and what is <.05 and why does that matter?
If you only have one IV, that is okay, but numerous IVs will artificially inflate the R2.
Need more? I am kind of tired so I will touch base tomorrow.

PS You have constraints!!!! Don't forget to talk about them, having a limit of 1k additional students is different than unlimited students.
  Reply With Quote
 
     
 
 
Old May 13, 2007, 01:41 PM   #3  
New Member
femaleinjax is offline
 
Join Date: May 2007
Posts: 2
femaleinjax See this member's comment history on his/her Profile page.
Quote:
Originally Posted by gogosean
Here is one they don't tell you a lot. The Y intercept, which is A, is sometimes considered Fixed Cost. The R squared evaluates % variance that the IV accounts for the DV, so model 2 is a better fit. I would also include the interpretation of the (coefficients/beta weights). Also, you have t values for all predictors, what is <.10 and what is <.05 and why does that matter?
If you only have one IV, that is okay, but numerous IVs will artificially inflate the R2.
Need more? I am kind of tired so I will touch base tomorrow.

PS You have constraints!!!! Don't forget to talk about them, having a limit of 1k additional students is different than unlimited students.


************************************************** ***************
Can you explain in more detail or provide me a website that shows detail steps?

Thanks
  Reply With Quote
 
     


Thread Tools
Display Modes

 
Similar Sponsors

Similar Threads
Question Asker Forum Answers Last Post
Linear Equations ray_mezz2002 High School 2 Jun 22, 2007 10:13 AM
statistics? with REGressIONS? bee1234 Math & Sciences 0 Apr 11, 2007 03:44 PM
LINEAR inequalities babygurl2323 Math & Sciences 3 Feb 5, 2007 09:29 PM
About linear...exponential highest Mathematics 0 Jan 20, 2007 09:35 PM
linear equtions quenta High School 4 Jan 3, 2007 06:42 AM




Copyright ©2003 - 2007, Ask Me Help Desk.
All times are GMT -8. The time now is 12:45 AM.