1
Sampling and sampling distributions.
Chapter 7.
So far, we have looked at simple techniques to display and
summarize data, and at probability as a measure of uncertainty.
Sample
Population
Describe the
sample data
Compute measures of
variability
•
Parameters
are numerical descriptive measures for
populations (fixed,
often unknown values)
For instance: For the normal distribution: the parameters
are location and shape , described by
µ
and
σ
.
For the binomial distribution : n and p.
•
Statistics
are numerical valued functions of the sample
observations.
(sample mean , sample variance).
• We use sample statistics to make inferences about the
corresponding population parameters.
•
Statistics vary from sample to sample and hence are
random variables.
Ò
Take 100 samples from the
middle class Americans and
record the age.
The average age of the population is
μ.
Population
X
1
…X
N
Sample 1,
35
x
1
=
Sample 2,
37
x
2
=
Sample 3,
34
x
3
=
Sample 4,
36
x
4
=
•The probability distributions for
statistics are called
sampling
distributions
.
•The sampling distributionthe
distribution of the possible values
of a statistic is the probability
distribution derived from
repeatedly extracting samples of
size n from the population.
•The sampling distribution depends
on a population distribution, sample
size n, the way you choose your
sample
Sampling variability
• Suppose I repeat the sampling 100 times. I will get different
mean values.
Values on the xaxes are the means computed in the 100 samples.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document2
N
x
μ
i
∑
=
where:
N = Population size , n = sample size
Sample Statistics are used to estimate population parameters
Problems:
–
Different samples provide different estimates of the population
parameter
–
Sample results have potential variability, thus sampling error exits.
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '08
 palmero
 Normal Distribution, Standard Deviation, Variance, C D Uniform Distribution

Click to edit the document details