Modeling and Prediction of Birth Rate in China

Abstract

Concerns about a persistent reduction are growing against the backdrop of large changes in Chinas birth rate during the last few decades. This paper explores this tendency via the lens of time series analysis, using birth rate records from 1964 through 2021 (particularly ignoring 1960-1963 due to Great Leap Forward distortions). The ARIMA and ETS models were extensively studied in our hunt for the best accurate forecasting device. The ARIMA (0,0,1) model was considered to be preferred based on comparison measures. The primary goal of this model was to predict the trend of Chinas fertility rates over the following five years. The ARIMA an1``d ETS models were rigorously applied to a selected training set after initial adjustments to ensure data stationarity, followed by an evaluation of their accuracy. Our findings, which are backed by the ARIMA model, imply a disturbing trend: a 0.117 percent annual fall in Chinas birth rate from 2022 to 2026. This suggests that a national fertility crisis is on the horizon. As a first step, we advise looking at the various socioeconomic reasons that may be driving this trend, as well as evaluating policy actions that could serve as potential cures.