Significance of regression, correlation coefficient, hypothesesi, confidence interval

silverjlp

Starting to get Involved
#1
Hello,

I am a statistics student. I have a Minitab assignment. My professor is making us learn Minitab on our own - really. I desperately need help and step by step help. I know very little about Minitab.


Here are the questions:
The weight and systolic blood pressure of 26 randomly selected males Assume that weight and blood pressure are jointly normally distributed.
(a) Find a regression line relating systolic blood pressure to weight.
(b) Test for significance of regression using α = 0.05.
(c) Estimate the correlation coeffiient.
(d) Test the hypothesis that ρ = 0, using α = 0.05.
(e) Construct a 95% confidence interval for the correlation coefficient.


Here is the data:
Weight Systolic BP
165 130
167 133
180 150
155 128
212 151
175 146
190 150
210 140
200 148
149 125
158 133
169 135
170 150
172 153
159 128
168 132
174 149
183 158
215 150
195 163
180 156
143 124
240 170
235 165
192 160
187 159

I think I can do the regression line using Stat - Regression - Fitted line plot, right?

How do I answer b-e?
You are not doing my homework for me, even if you answer step by step.
I appreciate all you can offer
 

Miner

Forum Moderator
Staff member
Admin
#2
Re: Significance of regression, correlation coefficient, hypothesesi, confidence inte

See my response to your other thread. Fitted Line Plot and regression will give you the same equations. The difference is that Fitted Line Plot will only handle one predictor, does provide a graph, but provides a condensed report in the Session Window. Regression can handle more predictors, including discrete, does not provide a graph, but provides more detailed reports.
 

silverjlp

Starting to get Involved
#3
Re: Significance of regression, correlation coefficient, hypothesesi, confidence inte

Hello
I am able to get a-c done, but I'm very lost on testing the hypothesis.

r, rho, is the population correlation coefficient for the paired variables

I can use STAT>BASIC STATISTICS>CORRELATION to find the values but I don't know how to test the statistic.
 

Miner

Forum Moderator
Staff member
Admin
#4
Re: Significance of regression, correlation coefficient, hypothesesi, confidence inte

r, rho, is the population correlation coefficient for the paired variables

I can use STAT>BASIC STATISTICS>CORRELATION to find the values but I don't know how to test the statistic.
When you use STAT>BASIC STATISTICS>CORRELATION, check the box labeled "Display p-values". Then compare the p-value from the Session Window output against your alpha = 0.05.
 

silverjlp

Starting to get Involved
#5
Re: Significance of regression, correlation coefficient, hypothesesi, confidence inte

Hi,
So I don't need to do a t test?

Thanks
 

Miner

Forum Moderator
Staff member
Admin
#6
Re: Significance of regression, correlation coefficient, hypothesesi, confidence inte

No. The p-value for the correlation test is all you need to answer the question.
 

silverjlp

Starting to get Involved
#7
Re: Significance of regression, correlation coefficient, hypothesesi, confidence inte

Fantastic! You have been so awesome. I truly cannot thank you enough for ALL of your replies.

Thank you. :agree1:
 
#8
Hello,

I am a statistics student. I have a Minitab assignment. My professor is making us learn Minitab on our own - really. I desperately need help and step by step help. I know very little about Minitab.


Here are the questions:
The weight and systolic blood pressure of 26 randomly selected males Assume that weight and blood pressure are jointly normally distributed.
(a) Find a regression line relating systolic blood pressure to weight.
(b) Test for significance of regression using α = 0.05.
(c) Estimate the correlation coeffiient.
(d) Test the hypothesis that ρ = 0, using α = 0.05.
(e) Construct a 95% confidence interval for the correlation coefficient.


Here is the data:
Weight Systolic BP
165 130
167 133
180 150
155 128
212 151
175 146
190 150
210 140
200 148
149 125
158 133
169 135
170 150
172 153
159 128
168 132
174 149
183 158
215 150
195 163
180 156
143 124
240 170
235 165
192 160
187 159

I think I can do the regression line using Stat - Regression - Fitted line plot, right?

How do I answer b-e?
You are not doing my homework for me, even if you answer step by step.
I appreciate all you can offer
 
#9
Hello friend , you first put your data in Minitab then at tool tab select [stat > display statistics> correlation ] so you get correlation analysis whether your relation strong or not. You get your Pearson coefficient also. After that in [Stat>regression> fit line model ] you select variables and you get regression anysis. And on graph you get coefficient of determination value [R^2] . that value should be more than 80% which shows your model is appropriate to explain variability.
 

Top Bottom