Scribd
Upload
23 views

Business Decision Making II Sampling Distributions: Dr. Nguyen Ngoc Phan

1) Sampling distributions describe the distribution of sample statistics like sums and means. The distribution of the sum of n independent observations has a mean of nμ and standard deviation of √nσ, where μ and σ are the mean and standard deviation of the population. 2) The Central Limit Theorem states that for a sample size n ≥ 30, the distribution of the sample mean will be approximately normal even if the population is not normal. The approximation improves as n increases. 3) A sample proportion p̄ is an estimate of the population proportion p. For large n, p̄ will be approximately normally distributed with mean p and standard deviation of √(p(1-p)/n

Uploaded by

Uyen Thu
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
23 views

Business Decision Making II Sampling Distributions: Dr. Nguyen Ngoc Phan

1) Sampling distributions describe the distribution of sample statistics like sums and means. The distribution of the sum of n independent observations has a mean of nμ and standard deviation of √nσ, where μ and σ are the mean and standard deviation of the population. 2) The Central Limit Theorem states that for a sample size n ≥ 30, the distribution of the sample mean will be approximately normal even if the population is not normal. The approximation improves as n increases. 3) A sample proportion p̄ is an estimate of the population proportion p. For large n, p̄ will be approximately normally distributed with mean p and standard deviation of √(p(1-p)/n

Uploaded by

Uyen Thu
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Business Decision Making II

Sampling Distributions

Dr. Nguyen Ngoc Phan

Dr. Nguyen Ngoc Phan Sampling Distributions


Sampling Distributions

For n independent observations of a random variable X , the


sum of the observations is the random variable

Sn = X1 + · · · + Xn .

The distribution of Sn is a sampling distribution.


Example: A random variable X has the probability distribution:
xi 2 3
pi 2/3 1/3

Find the probability distribution of S2 = X1 + X2 .

Dr. Nguyen Ngoc Phan Sampling Distributions


If X is a random variable with mean µ and standard deviation
σ, then
Sn = X1 + · · · + Xn
√
has mean nµ and standard deviation σ n.
Example: Let X be the weight of a hot cross bun. X has
mean 80 g and standard deviation 5 g. S6 is the total weight
of a packet of 6 hot cross buns. Find the mean and standard
deviation of S6 .

Dr. Nguyen Ngoc Phan Sampling Distributions


Distributions of Sample Means

For n independent observations of a random variable X , the


mean of the observations is the random variable
X1 + · · · + Xn
Xn = .
n
If the random variable X has mean µ and standard deviation
σ, then the distribution of sample means Xn has mean µ and
σ
standard deviation √ .
n

Dr. Nguyen Ngoc Phan Sampling Distributions


Example: The life expectancy X of a certain brand of AAA
battery is known to have mean µ = 27 hours and standard
deviation σ = 3.25 hours. The batteries are sold in packets of
6. Let the random variable X6 be the mean life expectancy of
batteries in a packet.
1 The 6 batteries in a packet were test, and the number of
hours they lasted were:

25.3, 21.6, 27.75, 22.25, 35.5, 28.5.

What is the corresponding observed value of the random


variable X6 ?
2 What is the mean and standard deviation of X6 ?

Dr. Nguyen Ngoc Phan Sampling Distributions


The Central Limit Theorem

Suppose X is a random variable which is not necessarily


normally distributed. X has mean µ and standard deviation σ.
For sufficiently large n (usually n ≥ 30), the distribution of the
sample means Xn of size n, is√approximately normal with mean
µ and standard deviation σ/ n. The larger the value of n, the
better the approximation will be.
The results regarding to the normality of the sample means Xn
also apply to the sum Sn = X1 + · · · + Xn .
Note that if X is itself normally distributed, then Xn and Sn
will be normally distributed for any value of n.

Dr. Nguyen Ngoc Phan Sampling Distributions


The Central Limit Theorem allows us to use normal
approximation to calculate probabilities associated with Xn and
Sn .
Examples: Problems 24 and 29 in page 325.

Dr. Nguyen Ngoc Phan Sampling Distributions


Sample Proportions

We are frequently presented with estimates of population


proportions, often in the form of percentages. For example:
If an election was held tomorrow, 52% of the population
would vote for candidate A.
73% of company executives say they will not employ
smokers.
To estimate a population proportion p, we consider taking a
random sample. From this sample, we obtain a sample
proportion p̄ which is an estimate of p.

Dr. Nguyen Ngoc Phan Sampling Distributions


We have
X
p̄ =
n
where X is the number of successes in n trials, X ∼ B(n, p)
where p is the probability of success.
Hence, X has the mean np and standard deviation
p
np(1 − p). Therefore
r
1 1 p(1 − p)
E (p̄) = E (X ) = p, σ(p̄) = σ(X ) = .
n n n

Dr. Nguyen Ngoc Phan Sampling Distributions


Let X be the Bernoulli random variable such that X = 1 if the
selected element has the given characteristic, and X = 0
otherwise. If we take n observations of X , then the sample
mean
X1 + · · · + Xn
X¯n =
n
is actually the sample proportion p̄. Hence, we can apply the
Central Limit Theorem and state that:
The normality of p̄
For sufficiently large n, p̄ is approximately
r normally distributed
p(1 − p)
with mean p and standard deviation .
n
As a general rule, p̄ is approximately normally distributed if
np ≥ 5 and n(1 − p) ≥ 5.

Dr. Nguyen Ngoc Phan Sampling Distributions


Example: Problems 35 and 39, page 330.
Homework: Problems 26, 28 page 325, and 36, 40 page 331.

Dr. Nguyen Ngoc Phan Sampling Distributions

You might also like