Question

4. (10 marks) Classify each statement by circling TRUE or FALSE. i) A scatterplot is not an attribute. TRUE FALSE ii) An unbiased estimator is always preferable to a biased estimator. TRUE FALSE iii) The observed significance level measures evidence against the null hypothesis. TRUE FALSE iv) The Horvitz-Thompson estimator is unbiased given a probability sampling design. TRUE FALSE v) When sampling with replacement the sample average is a Horvitz-Thompson estimate. TRUE FALSE vi) We can quantify the sampling error by calculating all possible samples. TRUE FALSE vii) The exact value of an estimator's bias is calculable based on all possible samples. TRUE FALSE viii) The standard bootstrap confidence interval is invariant to any one-to-one transformation. TRUE FALSE ix) One purpose of bootstrap method is to approximate the sampling distribution of an attribute. TRUE FALSE Increasing the complexity of a statistical model results in an increase in its prediction power. TRUE FALSE

Answer 1

Answers :-

(i) TRUE

(ii) FALSE as quality of the estimator is also depand on variability of the estimator.

(iii) TRUE  because if observed value of test statistic > tabulated value then we reject null hypothesis

(iv) TRUE

(v) TRUE

(vi) FALSE

(vii) FALSE

(viii) TRUE

(ix) TRUE

(x) FALSE