Wachemo University

College of natural and computational sciences

Department of statistics

Model exam for exit exam for Statistical Inference

General directions

In the following multiple choice questions, click on the correct answer.

1. Which of the following is true about statistical inference?

A) It is an advanced form of descriptive statistics.
B) It is all about computing the value of estimators.
C) It does nothing with testing valid hypothesis.
D) It is a summary name of statistical estimation and hypothesis testing about
population parameters.
2. Which of the following term well describes the statement “Statistic whose calculated
value is used to estimate a population parameter, θ”?
A) Inference C) Estimation
B) Estimate D) Estimator
3. One of the following statements is true about statistical estimation?
A) An estimator can be described by the numerical value assigned random variables.
B) Estimation does nothing with computation of parameter values.
C) There are two types of estimations.
D) Point estimation is the result of interval estimation.
4. The type of estimation or the approximation which take the form of one numerical
value is said to be?
A) Estimator C) Point estimation
B) Confidence interval estimation D) Inference
5. Which of the following is not the property of a good estimator?
A) Sharpness C) Consistency
B) Precision D) Unbiasedness
6. Which of the following is not the general property of point estimators?
A) Completeness C) sufficiency
B) Asymptotically normal D) Skewness
7. Which of the following is not one of the methods of finding point estimation?
A) Method of moments C) Bayesian
B) Maximum likelihood D) Interpolation
8. Given a sequence of n observed variables, the form
∏ represents:
A) Moments C) Likelihoods
B) Estimands D) Moments generating function
9. An interval of plausible values for the parameter being estimated, where degree of
plausibility is specified by:
 A) Point estimation C) Error
B) Maximum likelihood D) Confidence interval
10. One of the following is not additional method estimation other than moments,
Bayesian, and maximum likelihood methods?
A) Minimum-distance method C) Minimum-Chi-square method
B) Least squares method D) Mean-deviation method
11. Given and are two estimators of
and , for each in which one of the
following is true?
A) is a pitman-closer estimator of than .
B) is a UMV estimator of than .
C) is a UMV estimator of than .
D) is a UMV estimator of .
12. Which of the following statement is not true among others?
A) Closeness is thought in terms of increasing sample size.
B) Consistency and asymptotic efficiency are defined in terms of increasing sample
sizes.
C) Properties defined for increasing samples are large sample properties.
D) Good estimators are not applicable in statistical estimations.
13. Given , which of the following is true about estimators?
A) represents mean-square-error.
B) represents unbiasedness.
C) gives an unbiased estimator.
D) represents a consistent estimator.
14. Which of the following is true about bias of estimator T?
A) - gives a bias of estimator T.
B) gives a bias of estimator T.
C) = forms a bias of estimator T.
D) can always generate unbiased estimators of T.
15. Given and which of the following is not true about
loss and risk functions?
A) is a mean-value-absolute deviation.
B) The value of some estimator is called a decision and its
function is a decision function.
C) The expression is said to be a loss function.
D) The expression is called a risk function.
16. Based on question number 15, which of the following statement doesn’t describe the
context well?
A) The loss can be interchangeably used as an error.
B) The measure of error can be given by the loss function.
C) The expression holds true.
D) The expression describes the error of the estimate.
17. Given the function an estimator of ,
shows:
A) Average error C) Risk function
B) Average loss function D) Trivial function
18. Which of the following function is not well formulated?
A) represents known mean squared error.
B) shows the absolute error.
C) represents squared-error loss function.
D) shows the average absolute-error risk function.
19. For given two estimators and , select the
statement which doesn’t hold true.
A) , for all in , then an estimator is a better estimator than .
B) If there is no better estimator, then an estimator is an
admissible.
C) For , an estimator is said to be a minimax estimator
for every t.
D) The expression stands for power risk function.
20. For the given sample , the concept or a function that tells us as much
about as the sample itself is given by:
A) Statistic C) Sufficient Statistic
B) Parameter estimation D) Efficient Statistic
21. For the given sample coming from the density , where is a
vector of parameters and if the conditional distribution of given
doesn’t depend on for any value of , then the statistic is
defined by:
A) a complete statistic C) a unique statistic
B) an efficient statistic D) a sufficient statistic
22. Based on question number 21, which of the following holds true?
A) The statistics , are said to be jointly sufficient for the conditional
distribution of given , doesn’t depend on .
B) The statistic may always not free of the parameter
C) The conditional distribution of given ,
doesn’t depend on if and only if the statistic unbiased estimator.
D) The statistic as the result is said to be complete and unique.
23. If the set of jointly sufficient statistics is a function of every other set of sufficient
statistics, this situation is defined by:
A) Sufficient statistic C) Minimal sufficient statistics
B) Efficient statistics D) Maximal sufficient statistics
24. A big class in which many of the parametric families of densities are considered is
called?
A) Gamma class C) Families of Weibull
B) Gaussian class D) Exponential family
25. Given the expression , what relational
result can be drawn?
A) holds true for all parameters.
B) , for .
C) , for with T unbiased estimator of

D) , for .
26. Based on question number 25 with which is an estimator of
τ(θ), which one of the following expression is given wrongly?
A) If , is a uniformly minimum-variance unbiased estimator
(UMVUE) of by considering other conditions.
B) For together with other condition, is a UMVUE of .
C) If the condition of completeness is satisfied together with other conditions, then
the estimator is said to be a UMVUE of
D) The condition only is enough to say an estimator is to be a
UMVUE of
27. Under the regularity conditions the formula with

represents:
A) Central limit theorem C) Cramer-Rao inequality
B) Cauchy-Schawarz inequality D) Rao and Blackwell
inequality
28. If we let be a random sample from by
taking , then one of the following doesn’t show the correct result?
A) C) The Cramer-Rao lower

bound=

B) D)
29. For the family of densities of T with the condition
, where is a statistic. Then, the family is said to be:
A) Consistent C) Complete
B) Unbiased D) Unique
30. Given a random sample taken from , which
one of the following expressions doesn’t represent the result of the UMVUE?
A) ∑ is complete and sufficient.
B) ̅ ∑ is a function of sufficient statistic.
C) ∑ is an unbiased estimator of .
D) ̅ is not UMVUE of .
31. Suppose we are given with the fixed data , the conditional density of
given , denoted by
is called:
A) Bayesian estimator C) Prior density
 B) Likelihood function D) Posterior density
32. One of the following statements doesn’t well describe the relationship about
frequentist and Bayesian approaches in statistical inference?
A) Frequentist assumes parameters are fixed quantities.
B) In Bayesian approach, parameters are random variables.
C) In Bayesian approach, prior belief matters while observed data in frequentist.
D) Both frequentist and Bayesian approaches are impractical world in statistics.
33. Which of the following can’t describe the Bayesian estimation well?
A) Posterior likelihood prior
B) The prior is the probability of the parameter and represents what was thought
before seeing the data.
C) The posterior represents what is thought given both prior information and the data
just seen.
D) There is no relationship between Baye’s theorem and this method estimation.
34. Which of the following is true about estimation?
A) Point estimators fall between a given intervals.
B) There is no problem of finding interval estimation.
C) There is no problem of determining the optimum interval estimators.
D) For and , a value ( ) of the
random interval ( ) is called a 100 % confidence interval of
35. Let . If Q has a distribution doesn’t depend on , then Q is
said to be:
A) full quantity function C) Pivotal quantity
B) Two-sided interval D) one-side quantity
36. Let be a random sample from , then which of the following one
is a pivotal quantity:
̅
A) ̅ C)
√
̅
B) ̅ D)
√
37. Which of the following one is among the methods of constructing confidence
intervals?
A) Pivotal quantity method C) Confidence region method
B) Algebraic method D) Hypothesis method
38. An assertion or conjecture about the distribution of one or more random variables is
called?
A) Statistical estimation C) Statistical hypothesis test
B) Statistical experiment D) Statistical control
39. A statistical hypothesis completely specifies the distribution, then it is called:
A) Complete C) Composite
B) Simple D) Unique
40. In hypothesis testing, the tentative assumption about the population parameter is
A) The alternative hypothesis
B) The null hypothesis
C) Either the null or the alternative
 D) None of these alternatives is correct.
41. In hypothesis testing if the null hypothesis is rejected,
A) No conclusions can be drawn from the test
B) The alternative hypothesis is true
C) The data must have been accumulated incorrectly
D) The sample size has been too small
42. The error of rejecting a true null hypothesis is
A) Type I error
B) Type II error
C) Is the same as 
D) Committed when not enough information is available
43. The probability of making a Type I error is denoted by
A)  C) 1 - 
B)  D) 1 - 
44. The probability of making a Type II error is denoted by
A)  C) 1 - 
B)  D) 1 - 
45. Your investment executive claims that the average yearly rate of return on the stocks
she recommends is more than 10.0%. You plan on taking a sample to test her claim.
The correct set of hypotheses is
A) H0 :  < 10.0% Ha:   10.0%
B) H0 :   10.0% Ha:  > 10.0%
C) H0 :  > 10.0% Ha:   10.0%
D) H0 :   10.0% Ha:  < 10.0%
46. The school's newspaper reported that the proportion of students majoring in business
is more than 30%. You plan on taking a sample to test the newspaper's claim. The
correct set of hypotheses is
A) H0: P < 0.30 Ha: P  0.30
B) H0: P  0.30 Ha: P > 0.30
C) H0: P  0.30 Ha: P < 0.30
D) H0: P > 0.30 Ha: P  0.30
47. If a hypothesis is rejected at the 5% level of significance, it
A) will always be rejected at the 1% level
B) will always be accepted at the 1% level
C) will never be tested at the 1% level
D) may be rejected or not rejected at the 1% level
48. For the test of if the conditions and
are satisfied for which , then the type of test is said
to be:
∏
A) Most Powerful Test C) ∏
B) Neyman Pearson Lemma test D)
49. Most powerful test for simple hypothesis, __________for the composite hypothesis?
A) Most powerful test C) Composite test
B) Generalized likelihood-ratio test D) Single test
50. Given random samples from each of two normal populations with means and
variances ( ) and ( ), which of the following hypothesis about the two
variances is not correctly stated?
A)
B)
C)
D)