Problem Set 1

Problem Set 1
This exercise is based on material found in Section 1. The questions are about distributions, sample
mean and variance, hypothesis testing and properties of estimators.
Qs 1, 2, 3 and 4 are about the normal distribution, how it can be expressed in different
ways and how to calculate probabilities from it. Qs 1 and 2 are similar with 2 being slightly
more difficult. Qs 3 and 4 are the same apart from the variance value, and this allows you
to see how the variance affects the probability.
1. Let X ~ N ( ,  2 ) . Which of the following statements is true?
(a) X +  ~ N (0,  2 ) ;
X −
(b) ~ N (0,1) ;
2
X − 2
(c) ~ N (0,1) ;

X −
(d) ~ N (0,1) ;

(e) None of the above.
2. Let X ~ N ( ,  2 ) . Which of the following statements is true?
X −
(a) ~ N (0,1) ;
2
X −
(b) ~ N (0,1) ;

(c) X −  ~ N (0,  2 ) ;
(d) Both (a) and (b) are correct;

(e) Both (b) and (c) are correct.
3. If the random variable X is normally distributed with a mean (or expected value) of 132 and a
variance of 25, calculate the probability that X will take on
(i) a value greater than 140;
(ii) a value greater than 130;
(iii) a value less than 135 but greater than 125.
4. Consider exactly the same set up as in Q3 but let the variable have variance of 16.
Qs 5, 6 and 7 involve testing hypotheses about the population mean, using both the t-
statistic and the confidence interval approach.
5. Given the following random sample from a normally distributed population
5, 8, 11, 6, 15, 9, 12, 10, 7, 5, 2, 6

(a) calculate the mean and variance for this sample;
(b) use both the confidence interval and test of significance approach to test the null hypothesis
H0 :  = 9 against the alternative H1 :   9 at the 5% level of significance, where  is the
population mean. Do we reject the null hypothesis?
6. A random sample of size 100 is taken from a normally distributed random variable X, where
X represents the marks from an A-level exam taken by all British students, and gives the
following sample mean and sample variance: X =68 and S 2 =8.5. Test the null hypothesis that
H 0 :  = 70%
H1 :   70%
at the 5% significance level, where  is the national average for this exam.
7. For a sample of data of size 35, for which the sample mean is 19.82 and the sample variance
is 6.45, which of the following statements about the population mean 𝜇 is true?
(a) The 95% confidence band (to 3 decimal places) for  is (18.948,20.692) and hence the
null H 0 :  = 20 versus H 0 :   20 cannot be rejected at the 5% level;
(b) The 95% confidence band (to 3 decimal places) for  is (18.948,20.692) and hence the
null H 0 :  = 20 versus H 0 :   20 is rejected at the 5% level;
(c) The 95% confidence band (to 3 decimal places) for  is (19.094,20.546) and hence the
null H 0 :  = 20 versus H 0 :   20 cannot be rejected at the 5% level;
(d) The 95% confidence band (to 3 decimal places) for  is (19.094,20.546) and hence the
null H 0 :  = 20 versus H 0 :   20 is rejected at the 5% level;

Qs 8 and 9 are about the properties of estimators. Q9 is the most difficult question in this
exercise. It requires you to use the Technical Appendix to Section 1 which states some of
the properties of the expectations and variance operators. Understanding how to apply
these in the fairly simple context of the sample mean estimator will aid in the
understanding of its application in regression.
8. An estimator ˆ of a parameter  is a Best Linear Unbiased Estimator if
(a) it has the largest variance amongst all unbiased estimators of  ;

(b) it has the lowest variance amongst all biased estimators of  ;
(c) it has the largest variance amongst all biased estimators of  ;
(d) it has the lowest variance amongst all unbiased estimators of  ;
9. (a) What is meant by an "unbiased estimator"? Show that the sample mean is an unbiased
estimator.
(b) Derive the variance of the sample mean estimator.
(c) What is meant by "best" when an estimator is said to be a "best linear unbiased estimator"?
What is the sampling distribution of the sample mean X , if n is large?
(d) Another estimator of the population mean is the sample median, which has sampling
( )
distribution X med ~ N  , 2 n , where n is the sample size,  is the mean of X and  2 is the
2
variance of X. Show that X is more efficient than X med .

Problem Set 1

Uploaded by

Copyright:

Available Formats

Problem Set 1

Uploaded by

Document Information

Original Title

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

Problem Set 1

Uploaded by

Copyright:

Available Formats

Problem Set 1

1. Let X ~ N ( ,  2 ) . Which of the following statements is true?

2. Let X ~ N ( ,  2 ) . Which of the following statements is true?

(d) Both (a) and (b) are correct;

(i) a value greater than 140;

(ii) a value greater than 130;

(iii) a value less than 135 but greater than 125.

5. Given the following random sample from a normally distributed population

5, 8, 11, 6, 15, 9, 12, 10, 7, 5, 2, 6

(e) None of the above.

8. An estimator ˆ of a parameter  is a Best Linear Unbiased Estimator if

(a) it has the largest variance amongst all unbiased estimators of  ;

variance of X. Show that X is more efficient than X med .

You might also like