Quantitative Methods
Sampling Techniques and the
Central Limit Theorem
Probability Models for Portfolio Return and Risk
Probability and Nonprobability Sampling
Probability sampling: sampling when we know the probability in
the population of each sample member
(Simple) random sampling: every population member has an equal
probability of being selected; systematic sampling selects every
nth population member to form an approximately random sample
Nonprobability sampling: use judgment of researcher, or low-cost
or readily available data to select sample items
© Kaplan, Inc. 1
1
Probability Models for Portfolio Return and Risk
Sampling Error
Sampling error: difference between a sample statistic and the true
population parameter (e.g., x – μ)
Nonprobability sampling: may lead to greater sampling error than
probability sampling
© Kaplan, Inc. 2
Probability Models for Portfolio Return and Risk
Stratified Random Sampling
Stratified random sampling uses a classification system to separate
the population into smaller groups using distinguishing characteristics.
1. Create subgroups from the population based on important
characteristics.
2. Select samples from each subgroup in proportion to the size
of the subgroup.
Match the characteristics of the sample distribution to that of the
underlying population so it is a representative sample.
© Kaplan, Inc. 3
2
Probability Models for Portfolio Return and Risk
Cluster Sampling
The overall population is divided into subsets (clusters).
Example: State-level data on the personal incomes of residents is
shown by county. The county is the subset (cluster).
One-stage cluster sampling: take random sample of clusters;
include all data from those clusters
Two-stage cluster sampling: select clusters and take random
samples from each
© Kaplan, Inc. 4
Probability Models for Portfolio Return and Risk
Nonprobability Sampling Methods
Convenience sampling: selecting sample data based upon its ease
of access and being readily available
Judgmental sampling: select observations from population based
on analyst’s judgment
© Kaplan, Inc. 5
3
Probability Models for Portfolio Return and Risk
Central Limit Theorem
For any population with a mean and variance 2, as the size of a
random sample gets large:
The distribution of sample means approaches a normal
distribution, with a mean and variance of σ2/n
Properties of central limit theorem
If the sample size is sufficiently large (n 30), sampling
distribution of the sample means is approximately normal
The mean of the population and mean of the distribution of all
possible sample means are equal
© Kaplan, Inc. 6
Probability Models for Portfolio Return and Risk
Standard Error of the Sample Mean
The standard error of the sample mean is the standard deviation
of the distribution of sample means.
When the population σ is known: X
n
s
When the population σ is unknown: sX
n
© Kaplan, Inc. 7
4
Probability Models for Portfolio Return and Risk
Standard Error of the Sample Mean
Example: A sample contains the past 30 returns for McCreary, Inc.
The mean return is 2%, and the sample standard deviation is 20%.
Calculate and interpret the standard error of the sample mean.
s 20%
sX sX 3.6%
n 30
Interpretation: For sample sizes of 30, the distribution of the sample
means would have a mean of 2% and a standard error of 3.6%.
© Kaplan, Inc. 8
Probability Models for Portfolio Return and Risk
Resampling
Two alternative methods of estimating the standard error of a
sample mean are the jackknife and bootstrap methods.
The jackknife calculates multiple sample means, each with one of the
observations removed from the sample. The standard deviation of
these sample means is used as an estimate of the standard error.
The bootstrap method draws and replaces repeated samples of size
n from the dataset. The standard deviation of these sample means is
our estimate of the standard error of the sample mean.
© Kaplan, Inc. 9