Hypothesis Testing Basics and Errors5 min
Hypothesis testing begins with stating the Null Hypothesis (H0) and the Alternative Hypothesis (Ha). H0 always contains the equality condition (=, >=, <=), while Ha represents the claim being tested (often inequality). The significance level (alpha) is the probability of rejecting H0 when it is actually true, known as a Type I error. A Type II error (beta) is failing to reject H0 when it is false. The power of a test (1 - beta) measures the ability to correctly reject a false null. There is a trade-off between Type I and Type II errors; reducing one increases the other unless the sample size grows.

Key Points

  • Null Hypothesis (H0) must contain the equality sign.
  • Type I Error: Rejecting a true H0 (False Positive).
  • Type II Error: Failing to reject a false H0 (False Negative).
  • Significance Level (alpha) = Probability of Type I Error.
  • Power of Test = 1 - Probability of Type II Error.
Tests of Means6 min
Tests concerning means use the t-distribution. For a single mean, the test compares a sample mean to a hypothesized population mean with n-1 degrees of freedom. When comparing two independent means, the 'Difference of Means' test often employs a pooled variance assumption with degrees of freedom equal to n1 + n2 - 2. For dependent samples, such as paired observations (e.g., returns of the same asset in different periods or matched pairs), the 'Mean of Differences' test is used. This focuses on the differences between pairs, effectively reducing the problem to a single mean test on the differences, with degrees of freedom equal to the number of pairs minus one.

Key Points

  • Single mean test uses t-stat with df = n - 1.
  • Difference of means (independent) uses pooled variance with df = n1 + n2 - 2.
  • Paired comparison test (dependent) analyzes differences (d) with df = n(pairs) - 1.
  • Paired tests eliminate variation caused by factors other than what is being tested.
Tests of Variances5 min
Tests concerning variance utilize Chi-Square and F-distributions. To test a single population variance against a hypothesized value, the Chi-Square statistic is used with n-1 degrees of freedom. This test is sensitive to the assumption of normality. To compare two population variances, an F-test is conducted. The F-statistic is the ratio of the two sample variances. By convention, for one-tailed tests, the larger variance is placed in the numerator to ensure the F-stat is greater than 1. The F-distribution is defined by two degrees of freedom values: one for the numerator and one for the denominator.

Key Points

  • Single variance test uses Chi-Square statistic ((n-1)s^2 / hypothesized_variance).
  • Chi-Square distribution is asymmetric and non-negative.
  • Difference in variances uses F-statistic (Variance1 / Variance2).
  • F-test has two sets of degrees of freedom (numerator and denominator).

Questions

Question 1

Which of the following statements best describes the Null Hypothesis (H0)?

View answer and explanation
Question 2

In a hypothesis test, what constitutes a Type I error?

View answer and explanation
Question 3

If a hypothesis test is conducted at a 5 percent significance level, what is the probability of a Type I error?

View answer and explanation
Question 4

How is the Power of a Test calculated?

View answer and explanation
Question 5

Which test statistic is appropriate for testing a hypothesis about a single population mean when the population variance is unknown?

View answer and explanation
Question 6

When testing the difference between two independent population means using the pooled variance method, what are the degrees of freedom?

View answer and explanation
Question 7

What is the primary advantage of a paired comparison test (dependent samples) over a test of independent means?

View answer and explanation
Question 8

Which probability distribution is used to test a hypothesis about a single population variance?

View answer and explanation
Question 9

In a test of the difference between two variances, how is the F-statistic typically constructed?

View answer and explanation
Question 10

If the p-value of a test is 0.03 and the level of significance is 0.05, what is the correct decision?

View answer and explanation
Question 11

What is the relationship between the probability of a Type I error and the confidence level?

View answer and explanation
Question 12

In a two-tailed test with a 5 percent significance level, where are the critical regions located?

View answer and explanation
Question 13

Calculate the degrees of freedom for a single mean t-test with a sample size of 24.

View answer and explanation
Question 14

You are testing if a fund's return is different from 1.2 percent. The sample mean is 1.4 percent, the hypothesized mean is 1.2 percent, the standard deviation is 3.8 percent, and the sample size is 24. What is the calculated t-statistic?

View answer and explanation
Question 15

When calculating the pooled variance for two independent samples, what assumption is made?

View answer and explanation
Question 16

For a test of difference between two means with sample sizes 400 and 500, what are the degrees of freedom?

View answer and explanation
Question 17

Which of the following describes a 'False Negative' in hypothesis testing?

View answer and explanation
Question 18

In a paired comparison test of returns for two indexes over 1200 days, what is the degrees of freedom?

View answer and explanation
Question 19

If a researcher wants to prove that a fund's return is greater than the benchmark, how should the hypotheses be stated?

View answer and explanation
Question 20

Calculate the Chi-Square statistic to test if the variance is 0.0016, given sample variance of 0.002 and sample size of 21.

View answer and explanation
Question 21

What happens to the critical t-value as the sample size increases (holding significance level constant)?

View answer and explanation
Question 22

For an F-test comparing two variances, variance A is 8 and variance B is 4. What is the F-statistic?

View answer and explanation
Question 23

In the context of the F-test for variances, what determines the two degrees of freedom values?

View answer and explanation
Question 24

Which test allows for checking if a population variance has changed after a specific event (e.g., regulation change)?

View answer and explanation
Question 25

If a researcher decreases the significance level from 5 percent to 1 percent, what is the impact on Type I and Type II errors?

View answer and explanation
Question 26

A test statistic falls into the rejection region. Which conclusion is correct?

View answer and explanation
Question 27

When calculating the standard error for a difference of means test, what role does the pooled variance play?

View answer and explanation
Question 28

A paired t-test results in a t-statistic of 2.598. The critical values are +/- 1.96. What is the conclusion?

View answer and explanation
Question 29

Which of the following is true regarding the Chi-Square distribution used in variance testing?

View answer and explanation
Question 30

In a one-tailed test checking if variance has increased, where is the rejection region located?

View answer and explanation
Question 31

If the hypothesized mean difference in a paired test is zero, what are we testing for?

View answer and explanation
Question 32

A calculated F-statistic is 0.8. What typically happens in this scenario during a manual calculation check?

View answer and explanation
Question 33

Which step comes immediately after 'State the decision rule' in the hypothesis testing process?

View answer and explanation
Question 34

What is the standard error of the sample mean calculated as?

View answer and explanation
Question 35

If a confidence interval for a mean is [1.0, 1.4] and the hypothesized mean is 1.2, what is the hypothesis test conclusion at the corresponding significance level?

View answer and explanation
Question 36

What does a p-value of 0.001 imply?

View answer and explanation
Question 37

For a two-tailed t-test with 23 degrees of freedom and a 5 percent significance level, the critical value is approximately 2.069. If the t-stat is -2.5, what is the decision?

View answer and explanation
Question 38

What implies that the 'Power of the test' has increased?

View answer and explanation
Question 39

In a Chi-Square test for variance, if the test statistic is 22.54 and the critical value is 13.091 for a left-tailed test (less than), what is the decision?

View answer and explanation
Question 40

What is the primary assumption for the 'Difference of Means' test using pooled variance?

View answer and explanation
Question 41

If sample size n=50, what is the degrees of freedom for a Chi-Square variance test?

View answer and explanation
Question 42

A researcher assumes that stock returns are normally distributed. This is an example of:

View answer and explanation
Question 43

In a 'Mean of Differences' (paired) test, the sample mean difference is 0.3 percent, standard deviation of differences is 4.0 percent, and n=1200. What is the approximate t-statistic?

View answer and explanation
Question 44

To test if two variances are different (two-tailed), which F-test critical values are needed?

View answer and explanation
Question 45

If a test statistic is t = 4.76 and the critical value is 2.069, what is the outcome regarding the Null Hypothesis?

View answer and explanation
Question 46

Why might a researcher choose a 1 percent significance level instead of 5 percent?

View answer and explanation
Question 47

What does the 'Mean of Differences' test specifically analyze?

View answer and explanation
Question 48

If degrees of freedom for the numerator is 120 and denominator is 60, which test is likely being performed?

View answer and explanation
Question 49

Which of the following represents a non-directional test?

View answer and explanation
Question 50

If a Null Hypothesis is 'Mean Returns = 0', which Alternative Hypothesis implies a two-tailed test?

View answer and explanation