In: Statistics and Probability
Construct 90%, 95%, and 99% confidence intervals to estimate μ from the following data. State the point estimate. Assume the data come from a normally distributed population.
12.1 11.6 11.9 12.6 12.5 11.4 12.0 11.7 11.8 12.1
(Round the intermediate values to 4 decimal places. Round your answers to 2 decimal places.)
90% confidence interval: _____ ≤ μ ≤ _____
95% confidence interval: _____ ≤ μ ≤ _____
99% confidence interval: _____ ≤ μ ≤ _____
The point estimate is _____ .
Solution :
We are given a data of sample size n = 10
12.1,11.6,11.9,12.6,12.5,11.4,12.0,11.7,11.8,12.1
Using this, first we find sample mean() and sample standard deviation(s).
=
= (12.1 + 11.6.......+ 12.1)/10
= 11.97
Now ,
s=
Using given data, find Xi - for each term.Take square for each.Then we can easily find s.
s = 0.3773
Note that, Population standard deviation() is unknown..So we use t distribution.
Our aim is to construct 90% confidence interval.
c = 0.90
= 1- c = 1- 0.90 = 0.10
/2 = 0.10 2 = 0.05
Also, d.f = n - 1 = 10 -1 = 9
= = 0.05,9 = 1.8331
( use t table or t calculator to find this value..)
The margin of error is given by
E = /2,d.f. * ( / n)
= 1.8331* ( 0.3773/ 10 )
= 0.2187
Now , confidence interval for mean() is given by:
( - E ) < < ( + E)
( 11.97 - 0.2187 ) < < ( 11.97 + 0.2187 )
11.75 < < 12.19
Required 90% Confidence interval is ( 11.75 , 12.19 )
Construct 95% confidence interval.
c = 0.95
= 1- c = 1- 0.95 = 0.05
/2 = 0.05 2 = 0.025
Also, d.f = n - 1 = 10 - 1 = 9
= = 0.025,9 = 2.2622
( use t table or t calculator to find this value..)
The margin of error is given by
E = /2,d.f. * ( / n )
= 2.2622* ( 0.3773/ 10 )
= 0.2699
Now , confidence interval for mean() is given by:
( - E ) < < ( + E)
( 11.97 - 0.2699 ) < < ( 11.97 + 0.2699 )
11.70 < < 12.24
Required 95% confidence interval is ( 11.70 , 12.24 )
Construct 99% confidence interval.
c = 0.99
= 1 - c = 1 - 0.99 = 0.01
/2 = 0.01 2 = 0.005
Also, d.f = n - 1 = 10 - 1 = 9
= = 0.005,9 = 3.2498
( use t table or t calculator to find this value..)
The margin of error is given by
E = /2,d.f. * ( / n )
= 3.2498* ( 0.3773/ 10 )
= 0.3877
Now , confidence interval for mean() is given by:
( - E ) < < ( + E)
( 11.97 - 0.3877 ) < < ( 11.97 + 0.3877)
11.58 < < 12.36
Required 99% confidence interval is ( 11.58 , 12.36 )