Hypothesis Testing Flashcards

1
Q

Variable?

A

1 variable
2 variables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

1 variable?

A

Normal distribution:
•One dataset - one sample t-test
•Two related datasets - paired sample t-test
•Two independent datasets - independent sample t-test

Discrete distribution:
•Two related datasets - Wilcoxon test
•Two independent datasets - Mann Whitney U test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

2 variables?

A

Continuous
•Linear regression

Categorical
•Chi-square test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Test H0?

A

One sample- u(mu) = mean
Paired sample- u(mu) = 0
Independent sample- u1(mu)=u2(mu)
•F test- V1=V2
Wilcoxon- sum of R+= sum of R-
Mann Whitney- U1=U2
Linear regression- B=0
Chi square- O=E

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Risk analysis?

A

p<alpha - reject H0
Reject H0 - Risk of Type I error(rejecting a good H0)

p>alpha - accept H0 Accept H0 - Risk of Type II error
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

When to use Fischers test?

A

E>1
80% of test E>5

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Non Parametric test for one sample?

A

Wald wolfowitz

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Non parametric test for one, related and independent samples?

A

One- Sign test
Related- Friedman test
Independent- Kruskal-Wallis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Idea of hypothesis testing?

A

Check plausibility of hypothesis by testing the sample data, basically checking likelihood of a parameter of population testing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Power of test?

A

Probability of reject H0 at certain alpha

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Type of test for normal distribution?

A

Parametric tests

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Difference between Mann Whitney U and Independent?

A

In Mann Whitney only the ranking process is required

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

When to use Welch test?

A

When V1 not equal to V2 like for F test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Test for two continuous variables?

A

Pearson correlation coefficient tells us closeness of relationship unlike spearman’s rank correlation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Test for two categorical variables?

A

Chi-square

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Test for one continuous and one categorical(with >2 categories)?

A

ANOVA

17
Q

Test for one continuous and one categorical (with 2 categories)?

A

T test

18
Q

Principle of ANOVA test?

A

test for differences of means by examining the amount of variation within the samples

19
Q

For contingency tables?

A

Small size- Fischers test
Big size- Chi square

20
Q

Tests that apply Chi-square test?

A

Goodness of fit
Homogeneity test
Test of independence (non measurable)

21
Q

What is GK test used for?

A

Unknown probability of random variable

22
Q

Type of variable for sign test?

A

Nominal or ordinal