MODULE 9
z
Basic Concepts
on Estimation
z
Learning Targets
Discuss the properties of a good estimator
Illustrate point and interval estimation
Distinguish between point and interval estimation
Enrichment:
- Identify a point estimator for the population mean
- Compute for the interval estimate for a given confidence level
z
DESCRIPTIVE STATISTICS vs
INFERENTIAL STATISTICS
DESCRIPTIVE STATISTICS
- process of using or analyzing those measures that
quantitatively describe or summarize features from a collection of
information
INFERENTIAL STATISTICS
- pertains to the process of drawing and making decisions
concerning a given population based on the data obtained from a
sample
- focuses on estimation and hypothesis testing
z
Estimation
A process used to calculate the proposed values for parameters
by using only a random sample from the population.
Being done because population parameters are usually
unknown and /or the population is infeasible to study
Note: The results from estimation may not always be accepted
since it is preferred that interpretations and generalizations must
be made using data from the entire population. Thus, it is
necessary to do hypothesis testing.
z
Estimate
(Estimator)
POINT ESTIMATE INTERVAL ESTIMATE
- refers to a single value that - gives a range of values within
best determines the proposed which the parameter value
parameter value of the possibly falls
population
z
Properties of a Good Estimator
UNBIASEDNESS CONSISTENCY EFFICIENCY
When the When the When the
expectation of all estimate estimate gives
produced a the smallest
estimates taken
relatively small variance.
from samples are standard
equal to the error/deviation
parameter being (possible
estimated. amount of error
of estimating a
population
parameter)
z
Point Estimation
A point estimate is any simple statistic that can be used to represent the
value of a parameter.
The goal is to find the “best” point estimate.
z
Interval Estimation
It is not always sufficient to estimate the value of a parameter using
a single point estimator because there is doubt about the accuracy
of the point estimate.
The knowledge of sampling distribution can be used to construct an
interval around the point estimate and state the degree of certainty
(confidence) that the population parameter is within that interval.
z
Interval Estimation
The interval of values that predicts the population parameter is called
an interval estimate or confidence interval. The degree of certainty
that the true parameter falls within the interval is referred to as the
confidence level.
Confidence levels correspond to probabilities (areas) associated with
the normal curve.
z
Interval Estimation
C = confidence level
α =1–C
C/2 C/2
α/2 α/2
z
Interval Estimation
z
Interval Estimation
90.1% confidence interval is
