Aiming at the problems of certain randomness and low prediction accuracy of single prediction, based on the minimum sum of square error criterion, combined with the GIOWA operator, a combination model of four special parameters with variable weight coefficients is proposed. Taking the data of the per capita disposable income of urban residents in Anhui Province from 2000 to 2020 as a sample, a combined forecasting model based on the sum of squares of errors and the GIOWA operator is constructed to predict the per capita disposable income of urban residents in Anhui Province in the next 5 years. Firstly, three single forecast models are applied to fit and forecast the per capita disposable income of urban residents in Anhui Province. Then the study uses the minimum sum of square error as the criterion, combined with the GIOWA operator to construct a variable weight coefficient combination model, and at the same time takes four special parameters for the GIOWA operator to obtain the corresponding combination prediction model. Finally, the GIOWA combined prediction model is used to predict the per capita disposable income of urban residents in Anhui Province in the next 5 years. The results show that the prediction effect of the variable weight coefficient combination forecasting model is better than that of the single prediction model; the per capita disposable income of urban residents in Anhui Province will continue to grow steadily in the next 5 years.