Univariate vs Multivariate – Normality Check
Univariate vs Multivariate – Normality Check. Using the Q-Q Plot (Graphical Method) For Univariate Data Steps includes, 1. Arrange the values in ascending order (Data > Sort) 2. Compute the ranks of the values in the dataset (using RANK.EQ function) Manually change the ranks of the set of values having the same rank. 3. Calculate the cumulative probability of the ranks of the dataset, formula below. Assigned or Fitted Cumulative Probability = ((i – 0.5)/n) Where n is the size (count) of the dataset, and i is the rank of data value. 4. Calculate the normal predicted cumulative probability for the dataset based on empirical or theoretical normal distribution density function (using NORM.DIST function). 5. Finally, plot a scatter plot with the Predicted Normal Cumulative Probability on the x-axis and the Assigned Cumulative Probability on the y-axis. 6. If t