Year: 2025
Type: conference-paper
Abstract: This research zeroes in on the digital economy level in Guangdong Province. By tallying the keywords in government work reports spanning from 2002 to 2024, the digital economy level is quantified and the resultant data is partitioned into training and test sets. The findings spotlight that the ARIMA model exhibits prominent superiority. It adeptly reconciles the goodness of fit and complexity, precisely seizes data idiosyncrasies, manages non-stationarity, and fuses long-term trends with short-term oscillations. The AR model is apt for stationary data with robust autocorrelation, manifesting inertial trends. The MA model excels in capturing short-lived random vacillations. The ARMA model prevails when data possess both autocorrelation and short-term fluctuations. This research not only enriches the gauging modalities of the digital economy but also proffers valuable references for policymakers. Looking ahead, future research could broaden the investigative scope, amalgamate multi-source data, and probe deeper into the applicability of models within the segmented arenas of the digital economy, thereby contributing to a more comprehensive understanding and more efficacious management of the digital economic landscape in Guangdong Province.
Authors: Jiaxin Wang, Ziyu Su, Shuze Yao, Kaixiang Luo
Language: English
Subfield: Strategy and Management
Domain: Social Sciences
Open Access status: gold
OpenAlex ID: W4411888408