Scribd
Upload
156 views3 pages

Problems On Testing of Hypotheses-1 Solution

1. The document presents 4 problems related to testing statistical hypotheses. Problem 1 involves constructing a confidence interval for an unknown parameter θ and calculating a p-value. Problem 2 involves testing hypotheses about the number of defective televisions using a sampling procedure. Problem 3 involves testing hypotheses about the proportion of colorful butterflies using a rule to reject the null hypothesis. Problem 4 provides a table of brand preferences across regions and asks to determine the coefficients in the chi-square test statistic formula.

Uploaded by

Anmol Arora
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
156 views3 pages

Problems On Testing of Hypotheses-1 Solution

1. The document presents 4 problems related to testing statistical hypotheses. Problem 1 involves constructing a confidence interval for an unknown parameter θ and calculating a p-value. Problem 2 involves testing hypotheses about the number of defective televisions using a sampling procedure. Problem 3 involves testing hypotheses about the proportion of colorful butterflies using a rule to reject the null hypothesis. Problem 4 provides a table of brand preferences across regions and asks to determine the coefficients in the chi-square test statistic formula.

Uploaded by

Anmol Arora
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 3

Assignment Problems on Testing of Hypotheses

1. A certain quality characteristic in a production process has the following

probability density function: Here is a real-valued unknown parameter. A single observation, X , is drawn at random from this distribution. (a) Consider a confidence interval of the form [ X 2c, X + c] for , where c is a suitable positive constant. What is the value of c if the confidence interval has confidence coefficient 95%? (b) In order to test the null hypothesis H 0 : = 1 against the alternative H : 1 , the following test rule is proposed: Reject H 0 if T = X 2 2 X is too large. If X = 3 , then using this test rule, what is the p-value?
(Y ou can use the fact : F(x) = P(X x) is given by F(x) =(1/2) exp(x - ) w hen x < =1 - (1/2) exp( - x) w hen x )

f ( x) = 1 e x p( | x | ) , 2

< x< .

Solution:

( a ) P[ X 2c < < X + c ] = P[ 2c < X < + 2c] = F ( + 2c ) F ( 2c) = [1 exp( 2c )] = 0.95 Find c by trial and error.

(b)
P value = P[T > observed (T ) | H 0 ] = P[ X 2 2 X > 3 | H 0 ] = P[( X 3)( X + 1) > 0 | H 0 ]
= P[ X > 3 or X < 1 | H 0 ] = P[ X > 3 | H 0 ] + P[ X < 1 | H 0 ] = exp( 2)

2. (a) A dealer in electronic consumer goods received a supply of six televisions of brand A and ten televisions of brand B. Out of the six televisions of brand A, an unknown number, M, are defectives. Similarly, the number of defectives among the ten televisions of brand B is 2M. In order to test the null hypothesis H 0 : M = 2 against the alternative H a : M > 2 , the following procedure is adopted: (I)From the six televisions of brand A, draw a random sample of size two without replacement. (II) From the ten televisions of brand B, draw a random sample of size two without replacement. (III) Reject H 0 if and only if both the televisions in any of the two samples are defectives. (a) What is the probability of type I error for the above test procedure? Also, find the probability of type II error for this procedure when M equals 3.
Suppose X (Y) represents the number of defectives of brand A (B) in the sample. Decision Rule : Reject H 0 if X = 2 or Y = 2 P{Rejectin g H 0 } = P{ X = 2} + P{Y = 2} P{ X = 2}P{Y = 2} When M = 2, P{X = 2} =
2 4 C2 C 1 6 2 = , P{Y = 2} = 10 2 = = 6 45 15 C 2 15 C2

Thus P{Type I Error} = P{Rejectin g H 0 | M = 2} =

43 225 3 6 C C 3 15 1 When M = 3, P{X = 2} = 6 2 = , P{Y = 2} = 10 2 = = 45 3 C 2 15 C2 P{Type II Error at M = 3} = P{Not Rejecting H 0 | M = 3} = 1 7 8 = 15 15

3. A biologist, conducting research on butterflies, collects butterflies in a region. There are two kinds of butterflies in the region, colorful and boring (that is, notso-colorful), their respective proportions being and (1 ) . The butterflies are collected one by one and at random till exactly r colorful butterflies are obtained. Let X be the number of trials required to achieve this. The population of butterflies in the region is large so that the successive trials may be assumed to be independent. In order to test the null hypothesis H 0 : = 1 / 2 against the alternative H 1 : < 1 / 2 , the following test rule is suggested. Reject H 0 if X > c , where c is a suitable constant. (a) Let r = 1 and c = 3. For what the above test has power 0.6? In case no such exists, just say No. (b) Find the p-value when r = 2 and X = 4
2

Solution (a):

Power at = P{Rejectin g H 0 when is a value under the alternativ e} number of butterflie s capture till a colourful one is obtained.
x -1

= P{X > 3 | } where X represents P{X = x | } = (1 - )

P{ X < 3 | } = + (1 ) Power at =1 { + (1 )} Given condition :1 { + (1 )} = 0.6 Find by trial and error.

(b)

P value = P{ X observed X | = 1 / 2} = P{ X 4 | = 1 / 2} Now for r = 2, P{X = x} =x -1C1 (1 ) x 2 2 , x = 2,3,.....

Using this find the P - value

4. The following table concerns the findings of a market survey on two competing brands A and B across three regions 1, 2 and 3. From each region, a random sample of size 100 was drawn and then the people in the sample were classified according to brand preference. For instance, out of the 100 people sampled from the first region, x prefer A to B and 100 x prefer B to A, and so on. Brand preference A B Total 1 Region 2
y

Total
20 x + y

x
100 x

100 y

80 + x y

100

100

100

120 180 300

With reference to the above table, it is known that the chi-square statistic for testing homogeneity across regions is of the form Chi-square statistic = 1 ( g1 x 2 + g 2 y 2 + g 3 xy + g 4 x + g 5 y + g 6 ), 12 where g1 ,..., g 6 are constants. What are the values of g1 ,..., g 6 ? Note expected frequencies and observed frequencies both involve x and y. Represent Pearsonian Chi-square in the above form and then by comparing the value of the coefficients one can determine the values of g1 ,..., g 6 .

You might also like