Confidence Intervals.ppt
CIVL 3103
Confidence Intervals
Learning Objectives – Confidence
Intervals
Define confidence intervals, and explain their significance to point
estimates.
Identify and apply the appropriate confidence interval for
engineering-oriented problems.
Introduction
• We have discussed point estimates:
– as an estimate of a success probability, p
– as an estimate of population mean, µ
• These point estimates are almost never exactly equal to the true
values they are estimating.
• In order for the point estimate to be useful, it is necessary to
describe just how far off from the true value it is likely to be.
Confidence Intervals
Since the population mean will not be exactly equal to the sample
mean, , it is best to construct a confidence interval around that
is likely to cover the population mean.
We can then quantify our level of confidence that the population
mean is actually covered by the interval.
x x
The Central Limit Theorem
Suppose we have a population described by a random var iable X with a mean and a
standard deviation . We place no restrictions on the probability distribution of X. It
may be normally distributed, uniformly distributed, exponentially distributed, it
doesn’t matter.
Suppose we now take random samples from this population, each with a fixed and
large sample size n. Each sample will have a sample mean X , and this X will not, in
general, be equal to the population mean .
After repeated samplings, we will have built a population of Xs . The Xs are
themselves random variables and they have their own probability distribution!
The Central Limit Theorem says that, as long as n is reasonably large,
X N µ ,
σ2
n
⎡
⎣
⎢
⎤
⎦
⎥
If σ
2 n is the variance o f the sampling distribution, then the standard deviation
is σ n . This is commonly referred to as the standard error of the mean.
Confidence Interval on a Mean
(n large)
x
f(x)
α/2 α/2
2α
σ⎛ ⎞
± ⎜ ⎟
⎝ ⎠
x z
n
where 2αz is the critical point corresponding to a tail area of 2α
An equation for the (1–α)×100% confidence interval on a mean:
This equation can be used as long as n 30, even if σ is unknown. ≥
Example
a. Compute a 90% confidence interval for µ when
σ = 3.0, = 58.3, and n = 25.
b. Compute a 99% confidence interval for µ when
σ = 3.0, = 58.3, and n = 100.
c. How large must n be for the width of the 99%
confidence interval to be less than 1.0?
�
x
�
x
What if n is small?
Student’s t Distribution
• As the sample size becomes smaller, the sample standard
deviation becomes an increasingly poor approximation of
the population standard deviation. The end result is that a
95% confidence interval computed using s instead of σ may
actually only contain the population mean 90% of the time,
or 85% of the time, or even less.
• William Gosset developed a new probability distribution,
which he called the t distribution, to describe the
probabilities associated with the statistic
t =
x − µ
s
n
What if n is small?
Student’s t Distribution
Figure 5.9
Confidence Interval on a Mean
(σ UNKNOWN, n small)
An equation for the (1–α)×100% confidence interval on a
mean:
x ± tα 2,n−1
s
n
⎛
⎝⎜
⎞
⎠⎟
tα 2,n−1 α 2Where is the critical point corresponding to a tail area of .
Student’s t Distribution
Upper critical values of Student’s t distribution with ν degrees of freedom
Probability of exceeding the critical value
ν 0.10 0.05 0.025 0.01 0.005 0.001
1. 3.078 6.314 12.706 31.821 63.657 318.313
2. 1.886 2.920 4.303 6.965 9.925 22.327
3. 1.638 2.353 3.182 4.541 5.841 10.215
4. 1.533 2.132 2.776 3.747 4.604 7.173
5. 1.476 2.015 2.571 3.365 4.032 5.893
6. 1.440 1.943 2.447 3.143 3.707 5.208
7. 1.415 1.895 2.365 2.998 3.499 4.782
8. 1.397 1.860 2.306 2.896 3.355 4.499
9. 1.383 1.833 2.262 2.821 3.250 4.296
10. 1.372 1.812 2.228 2.764 3.169 4.143
Example
An unconfined compression test performed on 15
concrete cylinders produced the following strength
results (in psi):
2670 2580 2400
2490 2640 2590
2440 2170 2410
2590 2730 2690
2730 2480 2360
Find a 95% confidence interval for the true average
strength of the concrete.
Confidence Interval on Differences
(σ1 and σ2 KNOWN)
( )
2 2
1 2
1 2 / 2
1 2
α
σ σ
− ± +x x z
n n
where 2αz is the critical point corresponding to a tail area of 2α
This relationship is exact if the two populations are normally distributed. Otherwise, the confidence
interval is approximately valid for large sample sizes (n1 ≥ 30 and n2 ≥ 30).
An equation for the (1–α)×100% confidence interval on a difference in means:
Example
Aluminum spars from two different suppliers are used in
manufacturing the wing of a commercial aircraft. You have
been asked to determine if the latest shipments from each
supplier are equally strong. From past experience, the
standard deviations of the tensile strengths are known to be
1.5 kg/mm2 for Supplier 1 and 1.0 kg/mm2 for Supplier 2
(who has tighter quality control). A sample of 12 spars from
Supplier 1 has a mean tensile strength of 87.6 kg/mm2 and a
sample of 10 spars from Supplier 2 has a mean tensile
strength of 72.5 kg/mm2. If m1 and m2 denote the true mean
tensile strengths for the two shipments of spars, find a 90%
confidence interval on the difference in mean strength, m1 –
m2.
Confidence Interval on Differences
(σ1 and σ2 UNKNOWN but equal)
If random samples of size n1 and n2 are drawn from two normal populations with
equal but unknown variances, a 100(1– )% confidence interval on the difference
between the sample means, 1 – 2 i s:
( )
1 21 2 / 2, 2
1 2
1 1
α + −− ± +n n px x t S n n
where Sp is a “pooled” estimator of the unknown standard deviation and
is calculated as:
( ) ( )2 21 1 2 2
1 2
1 1
2
− + −
=
+ −p
n s n s
S
n n
But this can only be used if both populations are normally distributed.
Example
The drying time of pavement marking paint is of concern to
transportation engineers. Of two such paints from a particular
manufacturer, it is suspected that yellow paint dries faster than
white paint. Sample measurements of the drying times of both
paints (in minutes) are given below.
White: 120, 132, 123, 122, 140, 110, 120, 107
Yellow: 126, 124, 116, 125, 109, 130, 125, 117, 129, 120
Find a 95% confidence interval on the difference in mean drying
times, assuming that the drying times are normally distributed and
the standard deviations of the drying times are equal.
Confidence Intervals on Paired Samples
But this can only be used if both populations are normally distributed.
d ± tα 2,n−1
sd
n
⎛
⎝⎜
⎞
⎠⎟
An equation for the (1–α)×100% confidence interval on for a paired sample: d
Example
The manager of a fleet of automobiles is testing two brands of radial tires. He
assigns one tire of each brand at random to the two front wheels of eight different
cars and runs the cars until the tires wear out. The tire lives (in miles) are shown
below. Assuming that the tire lives for both brands are normally distributed, find a
99% confidence interval on the difference in mean life.
Car 1 2 3 4 5 6 7 8
Brand 1 36,925 45,300 36,240 32,100 37,210 48,360 38,200 33,500
Brand 2 34,318 42,280 35,500 31,950 38,015 47,800 37,810 33,215
Confidence Interval on the Variance
If a random sample of size n is taken from a normally distributed population, a 100(1–
α)% confidence interval on the variance of the population is:
( ) ( )n s n s
n n
−
≤ ≤
−
− − −
1 12
2 1
2
2
2
1 2 1
2χ
σ
χα α/ , / ,
But this can only be used if the population is normally
distributed .
Here, χα / ,2 1
2
n− and χ α1 2 1
2
− −/ ,n are the upper and lower critical points of the chi-square
distribution with n-1 degrees of freedom. Because the 2 distribution is asymmetrical,
the upper and lower tails are not the same.
Example
The compressive strength of concrete is being tested by a civil engineer. He tests 12
specimens and obtains the following data:
2216 2225 2318
2237 2301 2255
2249 2281 2275
2204 2263 2295
Find the 95% confidence interval on the population variance.
Confidence Interval on Ratio of Variances (σ1
and σ2 UNKNOWN):
s1
2
s2
2
1
Fα
2
,v1 ,v2
≤
σ1
2
σ 2
2
≤
s1
2
s2
2
Fα
2
,v2 ,v1
where :
v1 = n1 −1;v2 = n2 −1
F1−α ,u,v =
1
Fα ,ν ,u
A 100(1–α)% confidence interval on the ratio of variances (assuming both populations
are normally distributed) is:
Example
The diameter of steel rods manufactured on
two different extrusion machines is being
investigated. Two random samples of sizes n1
= 15 and n2 = 18 were selected from the two
machines. The sample means and variances are
m1 = 8.73, s21 = 0.35, m2 = 8.68, s22 = 0.40.
Construct a 95% confidence interval on the
ratio of the population variances.