Analysis and Prognostication of Residents' Per Capita Disposable Income in Hubei Province using Time Series Prediction Methods

Abstract

To anticipate the fluctuations in per capita disposable income among Hubei Province inhabitants for the subsequent biennium, a dataset spanning from 2005 to 2022 was culled. Employed in this study were three distinct time series prognostication methodologies: Exponential Smoothing (Holt-Winter), Autoregressive Moving Average (ARMA), and Autoregressive Integrated Moving Average (ARIMA). These techniques were applied to envision the forthcoming trajectory of per capita disposable income for the province's residents. By computing diverse metrics to assess predictive discrepancieslike the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE)the effectiveness of the assorted models was gauged, culminating in the selection of the ARIMA model due to its superior performance. Capitalizing on this, approximations for per capita disposable income during 2023 and 2024 were extrapolated. The resultant prognoses project a sustained and noteworthy uptick in per capita disposable income for urban denizens of Hubei Province in the forthcoming biennial span. Ultimately, the findings were translated into actionable policy suggestions and deductions, rendering them highly pertinent for the dissection of Hubei Province's economic evolution.