The correlation coefficient is a statistical measure of the strength of the relationship between the relative movements of two variables. The values range between -1.0 and 1.0. A calculated number greater than 1.0 or less than -1.0 means that there was an error in the correlation measurement. A correlation of -1.0 shows a perfect negative correlation, while a correlation of 1.0 shows a perfect positive correlation. A correlation of 0.0 shows no linear relationship between the movement of the two variables.
The coefficient of determination, denoted R² or r² and pronounced "R squared", is the proportion of the variance in the dependent variable that is predictable from the independent variable(s).
We shall see how the data is related if we create a scatter plot in power bi
Scatter Plot in Power BI
1. Click the Scatter Plot visualization and add your columns. For e.g. Year Month, Count Days, and Amount to determine the relationship.
2. From the Analytics pane add a Trend Line
There definitely seems to be a correlation between days and amount, so now we will make the calculations to see if we are right.
Creating the Coefficient of Correlation
- Right click on the table and click New quick measure
- Select Correlation coefficient from the Calculations under “Mathematical operations”.
- Select the Category, Measure X, and Measure Y. These columns will match the dot plot we created earlier.
The Coefficient of Correlation will now be available in your table, and it’s ready for use.
Creating the Coefficient of Determination
In this case, a quick measure would be overkill. The Coefficient of Correlation is notated as the letter R. The Coefficient of Determination is R2.
Coefficient of Determination = [Coefficient of Correlation]2
We now have two statistics based on the data set that tells how and to what degree the variables are related.
As we can see, per our definitions above, both the Coefficient of Correlation and Determination are very close to 1. This means Days is certainly related to Sum of Amount and does a very good job of predicting how much we will spend given the number of days we do something.0
Simple Linear Regression analysis is quite useful and prevalent across many business cases. Combining it with Power BI can create powerful analytical capabilities.
Thanks :)
No comments:
Post a Comment