Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

Comparison of Linear and Quadratic Regression Models

The following data presents the growth in worldwide Internet usage from 1995 through 2011:

Year:

1995

1996

1997

1998

1999

2000

2001

No. of

users (in

millions):

 

16

 

36

 

70

 

147

 

248

 

361

 

533

 

Year:

2002

2003

2004

2005

2006

2007

2008

No. of

users (in

millions):

 

597

 

719

 

817

 

1018

 

1093

 

1319

 

1574

 

Year:

2009

2010

2011

No. of

users (in

millions):

 

1802

 

2013

 

2267

 

1.   Enter the data into L1 and L2 and create a linear model for the data:

a.   What is your linear model?___________________________

b.   What is the slope?  _____

c.    Interpret the slope within the context of the problem.

d.    What is the r-value? _______  What does this tell you about the linear relationship between x and y?

e.   What is the r2 value?  What does this tell you about the relationship between x and y?

f.     Do a scatterplot of Y1 and the data.  What does that tell you about the linear

model?

g.    Store the predictions in L3 and the Residuals in L4.

h.    Do a scatterplot of X-values vs Residuals.  When doing this plot, first Deselect Y1. Sketch the plot here.  What does this tell you about the linear model?

i.    Create a quadratic model:___________________________

j.    What is the R2 value?  What does this tell you about the relationship between x and y?

k.    Do a scatterplot of Y1 and the data.  What does that tell you about the quadratic model?

l.     Store the predictions in L3 and the Residuals in L4.

m.   Do a scatterplot of X-values vs Residuals.  When doing this plot, first Deselect Y1. Sketch the plot here.  What does this tell you about the quadratic model?

n.    Look at the Residual plot and use the Trace Key to find the ‘largest’ residual (it could be positive or negative.)  Go back to your Stat Editor and find the corresponding x-    value, y-value and predicted y-value.

o.    Use your model stored in Y1 to estimate the number of internet users for 2012.