STAT 218
Week 4
Test / Chapter 6
TEST – Chapters 15
CHAPTER 6 – ESTIMATION AND STATISTICAL INFERENCE
REVIEW STATEMENTS
Recall  comment #1
•
We estimate unknown population parameters by sample statistics
o
E.g: Means for quantitative variables
•
It is an estimate but it is not necessarily accurate
•
How do we know HOW ACCURATE it is?
•
How confident are we that our estimate is close?
•
How close is it?
Recall – comment #2
•
The standard deviation of the mean got smaller as the sample size increased.
•
Actually by square root of n
Consider the survey of the number hours per week you spent listening to music
•
Estimate the average based upon a sample of size n = 4
•
If we repeated it again, we would not get the same mean. The estimate changes
from sample to sample
•
The larger the sample, the more confidence we have to estimate the population
quantity
•
This leads to the concept of a confidence interval
Standard Error
•
The standard deviation of the mean is called the Standard Error
•
Sec section 5.3 for previous discussion
•
Standard error
= s / √n
•
The difference between the sample mean and the population mean (µ) is rarely
more than a few standard errors
•
Reliability or precision of the estimate of the sample mean as an estimator of the
population mean (µ)
Standard Deviation vs Standard Error
•
SD describes the dispersion of the data
•
SE describes the uncertainty of the mean of the data.
1
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
•
The Standard Error incorporates two factors that affect reliability of the estimate
of the mean
Inherent variability of the data
(s)
The sample size (n)
Draw picture of s vs st.error showing change as n increases
SOME CONVENIENT DEFINITIONS
An
estimat
e is a specific value or quantity obtained for a statistic such as the
mean.
An
unbiased estimator
is one that produces an estimate that is expected to be
equal to the population parameter
A
point estimate
is a single number used to estimate a population parameter.
An
interval estimate
is a spread or interval of values used to estimate a
population parameter
The
confidence coefficient
refers to the probability of correctly including the
population parameter in the interval that is being produced. The probability is
designated as
“1.0 – alpha” The level of confidence is referred to as a percentage
(1.0 – “alpha”) x 100%
Confidence Intervals
are the interval estimates based upon the confidence levels
and are know as
confidence limits
.
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '08
 staff
 Statistics, Normal Distribution, Standard Error, representative

Click to edit the document details