From the course: Python for Time Series Forecasting
Join today to access over 24,800 courses taught by industry experts.
Interpretation of metrics in business terms - Python Tutorial
From the course: Python for Time Series Forecasting
Interpretation of metrics in business terms
- [Instructor] We are getting 10 passengers of error for any given forecast on average using the metric root mean squared error. To interpret this value in percentages, we must learn the bell curve with the nomenclature. We've got sigma that represents the standard deviation, which has practically the same formula as the root mean squared error. If we visualize the residuals, we observe that they mostly follow the normal distribution with the bell curve. Most of the differences are around zero. But at the end of the day, the deterministic conclusion that we achieved before of having 10 passengers of errors for any given prediction is false, because the probability of having a value falling into one particular point is zero. We must talk about intervals. So one deviation to the left and to the right makes up 68.2%. For any given value that the model forecasts, the difference with the real value that will happen in the future will be minus 10 or plus 10 68.2% of the time. Now, if you…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.