In this paper, the market demand forecasting of Sci-tech Service is investigated via an improved GMDH algorithm. The total revenue of Sci-tech Service Industry is specified as the index of market demand based on the assumption that the former is positively correlated with the latter. In order to reflect the research object veritably, fifteen influence factors are selected as the main parameters of such sci-tech service industry at first. Then, an improved GMDH algorithm is developed by adding initial variables into each layer, which is helpful to construct the multi-variable prediction model. In addition, BP neural network is adopted to predict the influence factors involved in the multi-variable model. Finally, simulation results are presented to verify the effectiveness of the proposed method.
Research on forecasting the market demand of Sci-tech Service Industry based on improved GMDH algorithm
2016-08-01
117198 byte
Conference paper
Electronic Resource
English
Forecasting Air Service Demand
SAE Technical Papers | 1972
|Multiinput-Multioutput Type GMDH Algorithm Using Regression-Principal Component Analysis
British Library Online Contents | 1993
|Development of Wave Overtopping Formulas for Inclined Seawalls using GMDH Algorithm
Online Contents | 2019
|Forecasting transportation demand for the U.S. market
Elsevier | 2019
|