Lecture 20

Lecture 20
STAT 30301 – Probability & Statistics for

Business
We Already Know:
Two Types of Statistical Inference:
• Confidence intervals are generally used to estimate a parameter ( or p)
• Hypothesis tests are used to measure the strength of evidence the data provide for a claim
Four Steps of Hypothesis Tests:

1. State the hypotheses: H0 (null) and Ha (alternative)
2. Compute the test statistic: When X is Normal or n > 30 and When X is not strongly skewed, has no outliers
is known use: and is unknown use:
3. Find the p-value based on the test statistic and the sign in H a
and/or find the critical value based on and the sign in H a
4. State the conclusion by comparing p-value to or test statistic to the critical value; 2 parts:
• Technical conclusion states “reject” or “do not reject” AND .
• Contextual conclusion must have context and include: “evidence”, the parameter, the
population, and be in terms of the alternative hypothesis
p-value Review:
• p-value: The probability of getting a test statistic at least as extreme as the
test statistic computed (using observed data) if, in fact, H 0 is true.
• small p-values say that the observed result is unlikely to occur when H0 is true.
• large p-values say that the observed result is likely to occur by chance when H0 is true.
• Hypothesis Testing Rejection Rule: Using the p-value to draw a conclusion

• If p-value ≤ α then:
reject H0 at . Conclude the data provide enough evidence to support H a,
remember to always use the context of the data in your conclusion!
• If p-value > α then:
Do not reject H0 at α. Conclude the data do not provide enough evidence to
support Ha  use the context of the data!
Critical Value(s) Review:
• Critical value(s): the value/values on the horizontal axis of the z or t distribution
that is/are the “cut-off” for the test statistic to provide evidence against the H0.
• These values determined by the test statistic(z or t) and .
• Hypothesis Testing Rejection Rule: Using the critical value(s) to draw a conclusion
Common z critical values:
• If Ha has < sign , then the critical value is OR
Reject H0 at if the test statistic critical value.
• If Ha has > sign , then the critical value is OR

Reject H0 at if the test statistic critical value.
, then the critical values are and
• If Ha has sign
OR and
OR if the test statistic positive critical value.
Reject H0 at if the test statistic negative critical value
Conclude the data do not provide enough evidence to support Ha
• Otherwise, do not reject H0 at α.
Recall: Inferring about
Categorical
Categorical Variables:
variables: type of car, candy color, yes/no, grade level, etc.
To analyze categorical variables statistically, use a proportion:

where: p = is a parameter
AND
is a statistic
So, p is the actual proportion of successes in the population,

and is the point estimate of p, from the sample.
Recall: Inference about Population
Proportion,
The sampling distribution of : p:
Shape: Approximately Normal if np 5 and n(1 – p) 5 
Center: Since is an unbiased estimator, = p
Spread: It can be shown that the Standard Error of :
Using statistics to approximate these parameters:
is the point estimate of p (Textbook uses )
is the estimate of
Conditions for inference:

1. randomness: data are from a random sample
2. Normality: np 5 and n(1 – p) 5
Hypothesis Test about Population Proportion, p:
Conditions:
1. Randomness: data are from a random sample
2. Normality: np0 ≥ 5 AND n(1 – p0) ≥ 5
Hypotheses: Reminder: For H0, your textbook uses the
Hypothesized Value sign that is the opposite of Ha but it always
H 0 : p = p0 has an equal sign: or or just
H a: p ≠ p 0 (or Ha: p < p0) ; (or Ha: p > p0)

Test statistic: ^ − 𝑝0
𝑝
𝑧=
√ 𝑝 0 (1 − 𝑝 0 )
𝑛
Example: Hypothesis Test
for p
Your friend claims 50% of the beads in a container are red. A
random sample of 251 beads is selected, of which 107 are red.
Test your friend’s claim at = 0.10.
Define the parameter:
Check conditions:
State the hypotheses:

Calculate the test statistic:
Find the p-value:
Find the critical value:
State the conclusions:

Example 2: Hypothesis Test
about p:
A potato-chip producer has just received a truckload of potatoes from
its main supplier. If the producer determines that more than 8% of the
potatoes in the shipment have blemishes, the shipment will be
refused. A supervisor selects a random sample of 500 potatoes from
the truck. An inspection reveals that 47 of the potatoes have
blemishes. Carry out a hypothesis test at the α = 0.10 significance level.
What should the producer conclude?
Define the parameter:
Check conditions:
State the hypotheses:
Calculate the test statistic:
Find the p-value:
Find the critical value:
State the conclusions:

Lecture 20

Uploaded by

Copyright:

Available Formats

Lecture 20

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture 20

Uploaded by

Copyright:

Available Formats

Lecture 20

STAT 30301 – Probability & Statistics for

Four Steps of Hypothesis Tests:

• Hypothesis Testing Rejection Rule: Using the p-value to draw a conclusion

• If Ha has > sign , then the critical value is OR

To analyze categorical variables statistically, use a proportion:

So, p is the actual proportion of successes in the population,

Conditions for inference:

H a: p ≠ p 0 (or Ha: p < p0) ; (or Ha: p > p0)

Define the parameter:

State the hypotheses:

Find the p-value:

Find the critical value:

State the conclusions:

Find the p-value:

Find the critical value:

State the conclusions:

You might also like