Research On GDP Forecasting Models Of Heilongjiang Province Using Long Time-series Nighttime Light Data

Nighttime light remote sensing data is regarded as an important data that can be used to measure human socioeconomic activities.Because of the positive correlation between long time-series nighttime light data and economic development,long time-series nighttime light data can be used to construct GDP forecasting models to predict GDP.However,there are different temporal resolutions,measurement standards,and errors due to different sources of nighttime light data.How to eliminate the data errors and the differences between different data sources to construct a reliable long time-series nighttime light data is currently a major problem.In addition,different GDP forecasting models have different applicable conditions and forecasting accuracy.How to choose the optimal GDP forecasting model is another major problem.In this paper,data errors and differences are dealt with at first to obtain long time-series nighttime light data that can be quantitatively analyzed.Then,the advantages and disadvantages of different GDP forecasting models are analyzed and compared.The main contents of this paper are as follows.Applying the methods of polynomial regression and RSR(ridgeline sampling regression),the intercalibration errors of DMSP/OLS(U.S.Air Force Defense Meteorological Satellite Program/Operational Linescan System)nighttime light data are calibrated.Three criteria for assessing calibration effects are proposed.DMSP/OLS nighttime light data obtained by combining two calibration methods is optimized.The NPP/VIIRS(Suomi National Polar-orbiting Partnership/visible infrared imaging radiometer suite)nighttime light monthly data is summed and averaged to obtain NPP/VIIRS nighttime light annual data.For dealing with the differences between DMSP/OLS and NPP/VIIRS nighttime light data,two assumptions are proposed to unify the two nighttime light datasets.Three kinds of GDP forecasting models are modeled and analyzed.They are linear regression model commonly used in the field of remote sensing,ARIMA model of univariate GDP time series analysis commonly used in the field of econometrics,and ARIMAX model of multivariate GDP time series analysis based on long time-series nighttime light data.ARIMAXe model with exogenous variables and ARIMAXs model with seasonal factors are constructed as submodels of ARIMAX model.Nighttime light data is applied to time series analysis for the first time.Finally,this paper compares and assesses the prediction results of four GDP forecasting models on the GDP of Heilongjiang province in 2017.