Applied Economics & Statistics Topic 7: Analysis of Variance Flashcards
- How to compare variances, for example, to test whether two samples come from populations that have equal variances. - How to test the equality of 3 or more population means. To do this we use an ANOVA (ANalysis Of VAriance) test. - Both of these tests require use of an F test. This F statistic has an F distribution
1
Q
Which tests require the use of an F test?
A
- Comparing variances, for example, to test whether two samples
come from populations that have equal variances - Testing the equality of 3 or more population means (using an ANOVA (ANalysis Of VAriance) test)
2
Q
State the characteristics of the F distribution
A
- A continuous distribution.
- Cannot be negative.
- Asymmetric (positively (towards the left) skewed).
- The F is a family of distributions,
with two degrees of freedom pa-
rameters, one in the numerator,
one in the denominator. - Asymptotic, i.e. the curve ap-
proaches the x-axis but doesn’t
reach it - It looks like the Maxwell-Boltzamann Distribution Curve
3
Q
Is a table needed for F distribution? What for?
A
Yes, F distribution table: this table gives the value of the distribution where
5% is in the tail. Each α level gets its own table.
4
Q
Describe and explain how, generally, an ‘Equal Variances Test’ goes
A
