The abstract states that the GM(1,1) model has low precision for predicting violent non-stationary data series. To address this disadvantage, it proposes both a GM(1,1) improved model and a Verhulst improved model using metabolism methods.
Conditions: The comparison applies to violent changing or violent non-stationary data series.; The empirical evaluation uses traffic accident data from 1996 to 2009 of Heilongjiang Province.
The abstract states that the GM(1,1) model has low precision for predicting violent non-stationary data series. To address this disadvantage, it proposes both a GM(1,1) improved model and a Verhulst improved model using metabolism methods.
Conditions: The comparison applies to violent changing or violent non-stationary data series.; The empirical evaluation uses traffic accident data from 1996 to 2009 of Heilongjiang Province.