You are on page 1of 53

Unit 5 MTH302

Etc.
Examples on How to write Ho and H1
1
ERROR in Hypothesis
explanation with following real life situation

Examples
Following content as per textbook or reference book
Que: For the same significant or critical value ‘𝒛𝜶 ’ if ‘𝜶’ is the
level of significance
for a two tailed test then the level of significance for a single tailed
test will be equal to:
%
(a) 2𝛼 (b) 𝛼 (c) (d) 𝑁𝑜𝑛𝑒 𝑜𝑓 𝑡ℎ𝑒𝑠𝑒.
&

For the same significant or critical value ‘𝒛𝜶 ’ the level


of significance for a
𝜶
single tailed test 0 2 is half the level of significance for
𝟐
two tailed test (𝜶).
𝑃(−3 ≤ 𝑍 ≤ 3) = 0.9973,
𝑖. 𝑒. , 𝑃 (|𝑍| ≤ 3) = 0.9973
𝑃(|𝑍| > 3) = 1 − 𝑃(|𝑍| ≤ 3) = 0.0027
𝑃(−1.96 ≤ 𝑍 ≤ 1.96) = 0.95,
𝑖. 𝑒. , 𝑃 (|𝑍| ≤ 1.96) = 0.95
𝑃(|𝑍| > 1.96) = 1 − 𝑃(|𝑍| ≤ 1.96) = 0.05
𝑃 (|𝑍| ≤ 2.58) = 0.99 𝑃 (|𝑍| > 2.58) = 0.01
When we attempt to make decisions about the population on the basis
of sample information, we have to make assumptions or guesses about
the nature of the population involved or about the value of some
parameter of the population.

Such assumptions, which may or may not be true, are called statistical
hypotheses there is no significant difference between the sample
statistic and the corresponding population parameter or between two
sample statistics
null hypothesis H0.
a hypothesis of no difference

alternative hypothesis H1
hypothesis that is different from
(or complementary to) the null hypothesis

A procedure for deciding whether to accept or to reject a null


hypothesis (and hence to reject or to accept the alternative hypothesis
respectively) is called the test of hypothesis

insignificant difference

difference which is caused due to sampling fluctuations

significant difference
difference that arises either because the sampling procedure is not
purely random or that the sample has not been drawn from the given
population

This procedure of testing whether the difference between


a is significant or not is called the test of significance
large samples follows normal distributions

t is a statistic in large samples and it follows


a normal distribution with mean E (t),
The error committed in accepting H0, when it is false

respectively.
The interval within which the parameter is expected to lie is called the
confidence interval for that parameter

The end points of the confidence interval are called confidence limits or
fiducial limits.
OR
å
Tests for small samples (n<30)
Test 1
Test 2
Above note 2 is also known as paired t-test
F-test
OR
OR
OR
ILLUSTRATIONS
For calculation of mean

You might also like