Lesson 1 |Multivariate Descriptive Statistics Flashcards

1
Q

What is the Joint Absolute Frequency? How do I calculate it?

A

The Joint Absolute Frequency can be calculated by assessing the number of observations for specific combination A & B. It is a pure number. Ex. 80 total of non-smokers and at the same time gym goers.

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

What is the Empirical Correlation Coefficient?

A

The Empirical Correlation is a coefficient that indicates and calculates the streght and the direction of a linear association.

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

In a contigency table, when do we have complete unrelatedness?

A

Unrelatedness happens when the proportion between two variables stays the same across different combinations. Ex. a1/b1=a2/b2= etc.

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

How do I calculate the unrelatedness?

A

The Unrelatedness values in a contigency table can be calculated by multiplying the jointed values for the column and the row of the cell, multiplying them and dividing everything by the total number of observations.

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

What is the Marginal Frequency?

A

The Marginal Frequency can be calculated through absolute or relative values. It is the sum of values per each column and row of the table.

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

What is a conditional contigency table?

A

A conditional contigency tables takes the either the values of the columns or of the rows as the 100% in order to analyse specific proportions such like: How many man vs women voted specifically for the column CDU?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How do you calculate the values of the conditional contigency table?

A

The values of the conditional contingency table will depend on whether we will take the row or the columns as 100%. If you are interested into checking the percentage of a certain combination compared to the row (a1 or a2) then you have fx(b1|a1) and you calculate the observation/total number for that row. The same is valid reversed for the columns fx(a1|b1).

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

How to calculate the Empirical Correlation Coefficient? What is it composed of?

A

The Empirical Correlation Coefficient is calculated as the Covariance of X and Y divided by both the Standard deviation of Y and the Standard Deviation of X.

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