MCQ 27

51.    The least square estimates have maximum variances among all the linear unbiased biased estimates.

A.       True.

B.        False.

C.        No comments.

D.        Depending upon situation.

52.    The answer to why we need general linear regression model in deviated form is that, in phenomena of response and explanatory variables the researcher is always interested in the intercept parameters. We have interested only in regression co-efficients i.e. β₁ β₂ β_{3 ………………..} β_k.   Then the model can be reduced by taking deviation from mean. We estimate β₁ β₂ β_{3 ………………..} β_k from deviated model by the

A.        Method of moments.

B.        Method of maximum likelihood.

C.        Moments method.

D.        Method of least square.

53     At a time data collected from different commodities is called 

A.            Crossectional data.

B.             Time series data.

C.                      Primary data.

D.                     Secondary data.

54.    The data collected for one commodity at different times is called

A.           Crossectional data.

B.            Time series data.

C.                    Primary data.

D.                    Secondary data.

55.    Extraneous information is the information obtained from any other source outside the sample which is being used for

A.        Examination.

B.        Projection.

C.        Population.

D.        Estimation.

56.   The ……………………….. information may be available in the following cases.

a)        We may have information on the exact value of some parameter from institutional information eg in the taxation law on the cigarette consumption defines that taxation is 70% of the retail price of tobacco manufactures.

b)        We may have information on the exact linear relationship between parameters.

A.        General.

B.        Extraneous.

C.        Endogenous.

D.        Special.

57.      There are three methods through which extraneous information are utilized.

a)        Method of transformation of variable.

b)        Method of pulling crossectional data and time series data.

A.       Method of using prior information.

B.       Moment method.

C.       Method of maximum likelihood.

D.       Method of least square.

58.      Refers to the case in which two or more explanatory variables in the regression model are highly correlated, making it difficult or impossible to isolate their individual effects on the depended variable.

A.        Correlation Matrix.

B.        Multicollinearity.

C.        Collinearity.

D.        Bicollinearity.

59.      With multicollinearity , the estimated OLS coefficient may statistically 

A.        Significant.

B.        Insignificant.

C.        Least.

D.       Maximum.

60.      Multicollinerity can sometimes be overcome or reduced by collecting more data, by utilizing prior information, by transformating the functional relationship or by dropping one of the highly

A.        Correlated numerical.

B.        Correlated parameters.

C.        Correlated variables.

D.        Correlated values.                

61.      One of the possible sources of Multicollinearity is that the data collected method can lead to the problem of mulitcollinearity when the investigator samples only a subspace of the K- dimensional space of the predictor variable.

A.        Projected variable.

B.        Estimated variable.

C.        Unknown variable.

D.        Predictor variable.

62.      Some procedures can tell us about multicollinearity phenomena.

a)        Examination of correlation matrix.

b)        Variance inflation factor.-

A.        Gauss method.

B.        The Farrar-Glauber test.

C.        Test Theorem.

D.        Examination of rank matrix.

63.      In case of perfect multicollinerity , the least square estimates

A.        Does not exist.

B.        Exists.

C.        Some time exists sometimes does not.

D.        Becomes very high.

64.      In case of strong multicollinerity it is possible to get OLS estimators but their SEs tend to be very large as the degree of correlation between explanatory variable

A.        Decreases.

B.        Increases.

C.        Sometimes increases and sometime decreases.

D.        Remains constant

65.      Because of large SEs the confidence interval for the relevant population parameters tends to be larger, hence the probability of accepting a false hypothesis

A.        Decreases.

B.        Increases.

C.        Sometimes increases and sometime decreases.

D.        Remains constant

66.      For strong multicollinearity the coefficient of multiple determination may also be

A.        High.

B.        Very high.

C.        Very low.

D.        Low.

67.      In case of strong multicollinearity, the OLS estimator and S.E becomes very sensitive to slight change in the

A.        Population data.

B.        Primary data.

C.        Secondary data.

D.        Sample data.

68.      If the OLS assumption that the variance of the error term is constant for all the values of independent variables does not hold, we face the problem of

A.        Hetrocedasticity.

B.        Homocedasticity.

C.        Parocedasicity.

D.        Herocedastictiy.

69.      This leads to unbiased but inefficient estimates of the S.Es.

A.        Hetrocedasticity.

B.        Homocedasticity.

C.        Parocedasicity.

D.        Herocedastictiy.

70.      Following are the few consequences of using OLS in the presence of ……………………………

a)        The least square estimate will have no bias.

b)        Variance of OLS coefficient will be incorrect.

A.        Herocedasticity.

B.        Homocedasticity.

C.        Parocedasicity.

D.        Hetrocedasticity..

71.      Name of tests for hetrocedasticity.

a)        The Spearman Rank Correlation test.

b)        The Gold Fled –Quant test.

A.        Run test.

B.        Q-test.

C.        The Park test.

D.        Test statistic.

72.    May be defined as “Correlation between members of series of observations ordered in time or space”.

A.        Auto Correlation.

B.        Exact Correlation.

C.        Inexact Correlation.

D.        Simple Correlation.

73.      There is slight difference between Auto Correlation and Serial Correlation. Auto Correlation is used for relation between lag values of the same series, while Serial Correlation is used for the relationship between current values of one series with the lag values of other series.

A.        No comments.

B.        False.

C.        True.

D.        Depending upon situation.

74.      Sources of ……………………………

a)        Omitted explanatory variables.

b)        Miss specification of the mathematical form of the model.

c)         Interpolation in statistical observations.

d)        Miss specification of true random error.

A.        Correlation.

B.        Auto Correlation.

C.        Rank Correlation

D.        Simple correlation.

75.      When the disturbance term exhibits serial correlation, the values as well as S.Es of parameter of estimates are

A.        Affected.

B.        Not affected.

C.        Correlated.

D.        Not correlated.

51	B	52	D	53	A	54	B	55	D
56	B	57	A	58	B	59	B	60	C
61	D	62	B	63	A	64	B	65	B
66	A	67	D	68	A	69	A	70	D
71	C	72	A	73	C	74	B	75	A