In: Statistics and Probability
Use the data shown in the table. Replace each x-value and y-value in the table with its logarithm. Find the equation of the regression line for the transformed data. Then construct a scatter plot of
left parenthesis log x comma log y right parenthesis(log x,logy)
and sketch the regression line with it. What do you notice?
x |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
|
y |
828828 |
353353 |
174174 |
103103 |
121121 |
6464 |
6969 |
3232 |
Find the equation of the regression line of the transformed data.
log
yequals=nothing
log
xplus+nothing
(Round to three decimal places as needed.)
Let's used excel:
First enter the given dataset in excel column:
then using LN() this command to transfer the data in natural logarithmic form
Regression using Excel.
Step 1) Click on Data >>> Data Analysis >>>Regression >>>>OK
Step 3) Input Y Range: Select the data of column "log Y"
Input X Range: Select all the data from "log x " column.
Click on Lable
then Click on Ouput Range
Look the following Image
Then Click on OK, we get following result.
from the above output the regression line is:
loy y = 6.780 + - 1.434 log x
Scatter plot:
Select the data of both columns as log x and log y
then click on Insert >>>Scatter>>>select first image
then we get the following output:
Now right click on any point on the scatter plot
then select Add Trendline...
Select Linear and choose "Display equation on chart"
Look the following image:
Then click on Close
So we get the following output:
From the scatter plot we conclude that the correlation between log x and log y is approximately linear and negative.