2% is the rule of error. If it is more than 2% it is not normal. Or not the right population. Or if it is more than 2% in error then, your sample is all wrong. But I think that would be a measure of regression (assuming you can measure and know how to measure change). I think this is a linear behavior model, since 2% is steady as she goes. Normalized, averaged, with standards (assuming you have standards after some point of time and sampling, otherwise you wait and see).
🚀Sign up for Our Complete Data Science Training with 57% OFF: bit.ly/3sGBk7a
2% is the rule of error. If it is more than 2% it is not normal. Or not the right population. Or if it is more than 2% in error then, your sample is all wrong. But I think that would be a measure of regression (assuming you can measure and know how to measure change). I think this is a linear behavior model, since 2% is steady as she goes. Normalized, averaged, with standards (assuming you have standards after some point of time and sampling, otherwise you wait and see).