Statistical Decision Theory (Chapter -10)-

https://www.simplilearn.com/tutorials/statistics-tutorial/hypothesis-testing-in-
statistics#:~:text=Hypothesis%20testing%20is%20a%20statistical%20method
%20used%20to%20determine%20if,data%20to%20assess%20the%20evidence.

The extension to statistical decision theory includes decision making in the

presence of statistical knowledge (mean, mode)

Hypothesis testing or significance testing is a method for testing a claim or

hypothesis about a parameter in a population, using data measured in a sample.

Hypothesis testing is the method of testing whether claims or hypotheses

regarding a population are likely to be true.

Hypothesis testing is a statistical method used to determine if there is

enough evidence in a sample data to draw conclusions about a population.
It involves formulating two competing hypotheses, the null hypothesis (H0)
and the alternative hypothesis (Ha), and then collecting data to assess the
evidence.

Procedure of Testing a hypothesis:

1. Setting up hypotheses or State the hypotheses. There are two types of

hypothesis-
a. Null Hypothysis - ( H0 )-
i. The null hypothesis, stated as the null, is a statement about a
population parameter, such as the population mean, that is
assumed to be true.

ii. Null hypothesis is the hypothesis which is to be tested for

possible rejection under the assumption that it is true.
iii. Keep in mind that the only reason we are testing the null
hypothesis is because we think it is wrong.
b. Alternate hypothesis - ( H1 )
 i. the negation of Null hypothesis is called the alternate
hypothesis
ii. It is set in such a way that the rejection of null hypothesis
implies the acceptance of alternate hypothesis
2. level of significance- fix the level of significance
a. The level of significance is the maximum possibility of making a
TYPE-I error and is denoted by 𝞪 (alpha). I.e P[Rejecting H0 when H0
is true.]
b. The probability of making a correct decision is then (1-𝞪).
c. The best value for fixing the level of significance depends upon the
seriousness of the results of the types of error..
d. Commonly used level of significance is 5% (0.05) and 1% (0.01)
e. When 𝞪= 5%, this means that we are 95% confident that a
correct decision has been made.
f. For example, if an investigator wanted to reject the null only if
there was even stronger evidence than that ensured with
α=0.05, they could choose a =0.01as their level of significance.
3. Compute the test statistic - z
a. The test statistic is a mathematical formula that allows
researchers to determine the likelihood of obtaining sample
outcomes if the null hypothesis were true. The value of the
test statistic is used to make a decision regarding the null
hypothesis.
b. The value of the test statistic is used to make a decision in
Step 4
c. Most commonly used test statistics is z-distribution.
4. Make a decision-

TYPE-1 and TYPE -2 Error-

Type-1 - If we reject a hypothesis when it should be accepted-
Type-2 -> if we accept a hypothesis when it should be rejected
Critical Value - The value of the sample statistics that defines the regions of
acceptance and rejection is called critical values.

Lower Tail Test/ Left Tail test

When the mean in alternate hypothesis is less than some no, it is a left tail test.
Right Tail Test
When the mean in alternate hypothesis is greater than some no, it is a right tail
test.

A sample of 400 male students of a college is found to have a mean height of

171.38 cm. Can it be regarded as a sample from a large population with mean
height 171.17 cm and s.d. 3.30 cm.
Sol - n=400, mu= 171.17 and Xbar=171.38 cm, sigma=3.30
1. Null hypothesis : yes sample has been drawn from a large population with
mean height =171.17

2. Calculate the test statistics -

z=1.27
3. Level of sig=5%
4. Z alpha (.05)=critical value at alpha 0.05=-1.96 and 1.96
5. Decision-
TYPE I (ALSO KNOWN AS ‘α’) AND TYPE II (ALSO KNOWN AS
‘β’)ERRORS

Just like a judge’s conclusion, an investigator’s conclusion may be wrong.

Sometimes, by chance alone, a sample is not representative of the population. Thus
the results in the sample do not reflect reality in the population, and the random
error leads to an erroneous inference.

A type I error (false-positive) occurs if an investigator rejects a null hypothesis that

is actually true in the population;

a type II error (false-negative) occurs if the investigator fails to reject a null

hypothesis that is actually false in the population.

Although type I and type II errors can never be avoided entirely, the investigator
can reduce their likelihood by increasing the sample size (the larger the sample, the
lesser is the likelihood that it will differ substantially from the population).

False-positive and false-negative results can also occur because of bias (observer,
instrument, recall, etc.). (Errors due to bias, however, are not referred to as type I
and type II errors.) Such errors are troublesome, since they may be difficult to
detect and cannot usually be quantified.
Questions on Hypothesis Testing - Z- statistics

A type I error (false-positive) occurs if an investigator rejects a null hypothesis that is

actually true in the population;
a type II error (false-negative) occurs if the investigator fails to reject a null
hypothesis that is actually false in the population.
Questions for Practice - ( Mean)

1. the breaking strengths of cables produced by a manufacturer have a mean

of 1800 pounds and a standard deviation of 100 pounds. By a new
technique in the manufacturing process, it is claimed that the breaking
strength can be increased. To test this claim, a sample of 50 cables is tested
and it is found that the mean breaking strength is 1850 pounds. Can we
support the claim at the 0.01 significance level?
2. The claim is made that internet shoppers spend on the average $335 per
year. It is desired to test that this figure is not correct at α=0.075. Three
hundred internet shoppers are surveyed and it is found that the sample
mean = $354 and the standard deviation =$125. Find the value of the test
statistics, the critical value and give your conclusion.
3. A stenographer claims that she can type at the rate of 120 words per
minute. Can we reject her claim on the basis of 100 trails in which she
demonstrates a mean of 116 words and a standard deviation of 15 words?
Use 5% level of significance.
4. A sample of 900 members has a mean 3.4 cms and s.d 2.61 cms. Can the
sample be regarded as one drawn from the population with mean 3.25
cms? Using the level of significance as 0.05, is the claim acceptable?
5. The mean life time of a sample of 400 fluorescent light bulb produced by a
company is found to be 1570 hrs with a standard deviation of 150 hrs. Test
the hypothesis that the mean life time of the bulb produced by the
company is 1600 hrs against the alternate hypothesis that it is greater than
1600 hrs at 1% level of significance.

6. A random Sample of 100 students gave a mean weight of 58 kg with s.d. of

4 kg. Test the hypothesis that the mean weight in the population is 60kg.
7. An investor assumes that the mean of duty returns of a stock since
inception is greater than $10. The average of 30 days return sample is $9.9.
Assume the population s.d is 0.011. Can we reject the null hypothesis at 5
significance level.
8. A sample of 400 items is taken from a normal population whose mean is 4
and whose variance is also 4 (sd.=2). If the sample mean is 4.45, can the
sample be regarded as a truly random sample?
9. A new variety of potato grown in 250 plots gave rise to a mean yield of
82.7quintals per hectare with a s.d of 14.6 quintals per hectare. Is it
reasonable to assert that the new variety is superior in yield to the standard
variety with an established yield of 80.2 quintals per hectare?

Questions for Practice (Proportion)

1. A car manufacturer claims that 40% of all cars built by his concern will be still in running
condition after 10 years. A random sample of 400 cars built by his concern showed that
150 cars were still in running condition after 10 years. Test the claim at 1 % level.
2. The claim is made that 40% of tax filer use computer software to file their Taxes. In a
sample of 50, 14 used computer software to file their taxes. Test H0 : p=0.4 versus H1:
p<0.4 at α=0.05 where p is the population proportion who use computer software to file
their taxes.
3. A dice was thrown 9000 times and of these 3220 yielded a 3 or 4. Is this consistent with
the hypothesis that the dice was unbiased? Level of significance is 0.05. [hint: sample
proportion P=3220/9000 = 0.358, population proportion p=2/6=0.333, two tailed test. ]
4. In a sample of 400 burners there were 12 whose internal diameter were not within
tolerances. Is this sufficient evidence for concluding that the manufacturing process is
turning out more than 2% defective burners. Let α= 0.05.
[Hint p=0.02, q= .98, P= 12/400 = 0.03. Let Null Hypothesis is : the process is in control
i.e H0 : p<=0.02, H1 : p>.02, one tailed test]

5. A manufacturer claimed that at least 95% of the equipment which he supplied to a

factory confronted to specification. An examination of a sample of 200 pieces of
equipment revealed that 18 were faulty. Test this claim at a significance level of 0.05.
[Hint: p=0.95, q=0.05, sample proportion (P)=(200-18)/200= 182/200= 0.91, Null
Hypothesis = proportion of pieces confirming to specification is 95% i.e H0: p=0.95 and
Alternate Hypothesis = H1: p<0.95]