To validate the quality of forecasts and predictions of different quantities, this page generates statistics, comparing the predicted values with reference datasets.
The index quality statistics have been implemented with the kind support of Daniel Heynderickx from DH Consultancy and Petra Vanlommel from SIDC.
| Generated Statistics |
|---|
| Name |
Symbol |
Description |
| Correlation Coefficient |
r |
Is a measure of the reliability of the linear relationship between the x and y values. |
| Coefficient of Determination |
r2 |
Gives the proportion of the variance of one variable that is predictable from the other variable. |
| Mean Error |
ME |
Is a measure of the overall bias of the forecast. |
| Mean Square Error |
MSE |
A measure of the forecast accuracy. |
| Root Mean Square Error |
RMSE |
Is a frequently-used measure of the differences between forecast and observed values. |
| Forecast Skill Index |
S |
The average accuracy of a forecast method relative to a forecast produced with a reference method. |
| Linear Regression | The line with the minimized squared distance to all points. |
| Scatter Plot | Visual comparison of forecast vs. measured value for the last month. On a perfect forecast, the green line would match the blue line, such that the linear regression is 1 (x=y). The linear regression is the (green) line with the minimized squared distance to all points. |
| Box Plot | Shows the distribution of value deviations (forecast from measured values). Ideally, the greater part of the deviations would occur on the left hand side (smaller deviations). |
| Joint Distribution | The joint distribution shows the frequency of occurance of (forecast, observation) pairs. |
| Monthly Distribution | Box plot of the monthly distribution. Stars indicate the minimum and maximum values and the red line the mean observed values. The horizontal blue line within the boxes represents the median and the box encompass the 25th and 75th percentiles, while the box 'whiskers' encompass the 1st and 99th percentiles. |
| Annual Trend | Shows the annual trend of the mean observed values, the mean error and the root mean square error as well as the correlation coefficient. |
| Forecast Skill Index | The average accuracy of a forecast method over the year |
