Scribd
Upload
20 views29 pages

Lecture08 Hypothesis Testing Inf Stats FA24

The document discusses hypothesis testing in statistics, focusing on null and alternative hypotheses, Type I and Type II errors, and the significance of p-values. It explains how to evaluate the significance of evidence against a hypothesis using statistical tests and provides examples to illustrate these concepts. Additionally, it emphasizes the importance of understanding the implications of errors in hypothesis testing and the use of statistical tables for calculations.

Uploaded by

raoabdulraffay355
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
20 views29 pages

Lecture08 Hypothesis Testing Inf Stats FA24

The document discusses hypothesis testing in statistics, focusing on null and alternative hypotheses, Type I and Type II errors, and the significance of p-values. It explains how to evaluate the significance of evidence against a hypothesis using statistical tests and provides examples to illustrate these concepts. Additionally, it emphasizes the importance of understanding the implications of errors in hypothesis testing and the use of statistical tables for calculations.

Uploaded by

raoabdulraffay355
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 29

Inferential Statistics and

Applied Probability
Hypothesis Testing
Ch-9 from Michael Baron’s Book
Ch-7 from Forsyth’s Book

1
Introduction

2
Null and Alternative Hypothesis

• Let, we are interested in the “burning rate” of solid propellant in aircrew escape
systems which is a RV described by a probability distribution
• Let we are specifically interested in mean rate i.e. in deciding whether or not this
values is 50 cm/s. This can be formally expressed as:

• One-sided alternative hypothesis: Either μ<50 or μ>50


• Two-sided alternative hypothesis: μ≠50
• It is important to note that hypotheses are always statements about the population
or distribution and not about the sample
3
Null Hypothesis
• The value of population parameter specified in the null
hypothesis usually determined in one of three ways.
o First, it may result from past experience or knowledge of the
process, or even from previous tests or experiments. The
objective of hypothesis testing then is usually to determine
whether the parameter value has changed.
o Second, this value may be determined from some theory or
model regarding the process under study. Here the objective
of hypothesis testing is to verify the theory or model.
o A third situation arises when the value of the population
parameter results from external considerations, such as
design or engineering specifications, or from contractual
obligations. In this situation, the usual objective of hypothesis
testing is conformance testing.

4
Type I and Type II errors
• When testing hypotheses, we realize that all we see
is a random sample.
• Therefore, with all the best statistics skills, our
decision to accept or to reject H0 may still be wrong
i.e. may be due to a sampling error.
• Four situations are possible,

5
Type I and Type II errors (Cont)
• Each error occurs with a certain probability that we hope to
keep small. A good test results in an erroneous decision only
if the observed data is somewhat extreme.
• Type I errors more undesired e.g. convicting an innocent
defendant (null), hence their probability usually bound by a
pre-assigned small number α and it is endeavored to
minimize type II errors
• Type I error probability called Significance level of a test or
α-error
• The power of a statistical test is the probability of rejecting the
null hypothesis H0 when the alternative hypothesis is true

6
The P-value
• P-value is the probability that we would have seen
our data (or something more unexpected) just by
chance if the null hypothesis (null value) is true.
• Small p-values mean the null value is unlikely given
our data i.e. would result in null hypothesis being
rejected
• More formally, The P-value is the smallest level of
significance that would lead to rejection of the
null hypothesis H0 with the given data

7
The P-value (Cont)
• By convention, p-values <0.05 are often accepted as
“statistically significant” in the medical literature; but this is
an arbitrary cutoff.
• A cut-off of p<0.05 means that in about ONLY 5 of
100 experiments, a result would appear significant just by
chance (“Type I error”) so safe to use this value as threshold
to reject the Null hypothesis
o Null hypothesis is rejected when it is confirmed that something
unusual did not happen by chance but by evidence collected in
hypothesis testing
• It is customary to call the test statistic (and the data)
significant when the null hypothesis H0 is rejected

8
Explanation with Example

9
Example (Contd)

10
Summary: Hypothesis Testing

11
Significance
∘
• Let Null Hypothesis be that average body temp = 95 F
• Collect temp of random sample of∘ N people where the
sample mean is less likely to be 95 F
• We must find the reason for the difference between
the sample mean and our hypothesized value
• The hypothesis may be wrong or the difference may
just be due to samples being randomly chosen
• Significance of evidence against hypothesis assessed by
finding out what fraction of samples would give us
sample means like the one we observe if the hypothesis
is true.

12
Evaluating Significance
• Use t-statistic with N-1 degrees of freedom if N<30
or use Z-statistic if N>30

• For any observed value “s” of “S”, we get “t” i.e.


observed value of “T” usually called test statistic
• Fraction of samples with less extreme values

• Large value of f indicates most samples close to


zero i.e. hypothesis true, else its false
13
Fraction of less extreme samples

14
Rejection Region of Null
Hypothesis

15
P-value: Rejection region i.e.
Extreme Fraction

16
A Hypothesis

17
P-value: Election Polling

Use z-table and subtract P(z=-2.778)=0.00272 from P(z=2.778)=0.99728


and then substract this value from 1 i.e. 1-0.99456=0.0054
18
Z-Table

19
Z-Table

20
P-value Caution

21
Some Problems From Forsyth
Example: 7.1
• Assume the mean weight of a male chow eating
mouse is 35 gr. and the standard error of a sample
of 44 such mice is 0.827 gr. What fraction of
samples of 44 such mice will have a sample mean
in the range 33–37 grams?

22
Solution: Exp 7.1
𝑠−35
• The statistic to use is 𝑇 =
0.827
• The question is asking for the probability that t
takes a value in the range

33 − 35 37 − 35
, 𝑖. 𝑒. −2.41, 2.41
0.827 0.827
• So,

23
Z-Table

24
Z-Table

25
Example 7-2
• Assume the population mean of the weight of
a chow-eating female mouse is 27.8 gr and stderr
of 0.70. Estimate the fraction of samples that will
have mean weight greater than 29 gr. Consider
sample size of 48.
• Try!

26
Ex 7-2: Solution

27
Z-Table

28
Acknowledgements
• Some parts taken from the StatQuest youtube
channel
• Next time
o More significance tests e.g. F-test, chi-squared test etc

29

You might also like