Measurement of the relationship
1. Covariance
- Covariance is a way to classify three types of relationships between two continuous variables.
- Only the sign(positive or negative) matters.
2. Pearson's Correlation
-
Correlation tells you how strong two quantitative variables are related.
-
It ranges from -1 to +1
-
The maximum value for correlation, 1, occurs whenever you can draw a straightline with a positive slope that goes through all of the data.
-
The confidence in how useful the relationship is, determined by p-value, depends on how much data we have.
-
The more data, the smaller p-value.
Coefficient of Determination(R-squared)
- R-squared is the percentage of variation explained by the relationship between two variables.
- It is simply squared version of R.
- This means that most of the variation in the data is explained by the size/weight relationship.
- If the value is small(1%), it means something else must explain the remaining 99%.
yozzum