I have 2 different columns X1, and X2. As you can see scale of X1 is larger than X2.
These 2 columns represent the predicted value of performance of my two products. However the range or scale of X1 is higher than others because it is computed from different prediction model and different prior expectation value. But does not mean that it has better performance.
Note: X2 has some outliers value such as 0.91, which makes normalization methods more complex.
I tried scale() in R: as
X1scaled <- scale(X1)
to z-transform X1 and aftherwards multiply every value with sd(X2) and afterwards add mean(X2).
Now how to make sure my result is correct