Abstract:

Aiming to the inperfect adaptability of the traditional grey Verhulst model, using discretization thought for reference, a grey discrete Verhulst model is constructed by reciprocal generation of the original data sequence. The grey discrete Verhulst model gives full consideration to the quasi exponential law, realizes changes from continuous form to discrete form, and eliminates jumping errors from the differential equation to the difference equation. The prediction formulas of the grey discrete Verhulst model are given under two initial conditions, which effectively solves the traditional problem of low predicted stability. The final adoption of an example shows that the model can significantly improve simulation and prediction accuracy.