Data Science Questions and Answers Part-26

1. The binomial random variables are obtained as the sum of iid Gaussian trials.
a) True
b) False

Answer: a
Explanation: The binomial random variables are obtained as the sum of iid Bernoulli trials.

2. Which of the following testing is concerned with making decisions using data?
a) Probability
b) Hypothesis
c) Causal
d) None of the mentioned

Answer: b
Explanation: The null hypothesis is assumed true and statistical evidence is required to reject it in favor of a research or alternative hypothesis.

3. Point out the correct statement.
a) Power of a one sided test is lower than the power of the associated two sided test
b) Power of a two sided test is greater than the power of the associated one sided test
c) Hypothesis testing is less commonly used
d) None of the mentioned

Answer: d
Explanation: Power of a one sided test is greater than the power of the associated two sided test.

4. Which of the following value is the most common measure of “statistical significance”?
a) P
b) A
c) L
d) All of the mentioned

Answer: a
Explanation: The P-value is the probability under the null hypothesis of obtaining evidence as extreme or more extreme than would be observed by chance alone.

5. What is the purpose of multiple testing in statistical inference?
a) Minimize errors
b) Minimize false positives
c) Minimize false negatives
d) All of the mentioned

Answer: d
Explanation: A false positive is an error in some evaluation process in which a condition tested for is mistakenly found to have been detected.

6. Point out the wrong statement with respect to FDR.
a) FDR is difficult to calculate
b) FDR is relatively less conservative
c) FDR allows for more false positives
d) None of the mentioned

Answer: a
Explanation: FDR stands for false discovery rate.

7. Which of the following is the oldest multiple testing correction?
a) Bonferroni correction
b) Bernoulli correction
c) Likelihood correction
d) All of the mentioned

Answer: a
Explanation: Bonferroni correction is easy to calculate.

8. The pooled estimator is a mixture of the group variances, placing greater weight on whichever has a larger sample size.
a) True
b) False

Answer: a
Explanation: If the sample sizes are the same the pooled variance estimate is the average of the group variances.

9. Which of the following tool is used for constructing confidence intervals and calculating standard errors for difficult statistics?
a) baggyer
b) bootstrap
c) jacknife
d) none of the mentioned

Answer: b
Explanation: The bootstrap procedure follows from the so called bootstrap principle.

10. Which of the following tool is used for estimating standard errors and the bias of estimators?
a) knitr
b) jackknife
c) ggplot2
d) all of the mentioned

Answer: c
Explanation: jackknife involves resampling data.