关键词: 旅居车；销量预测；时间序列；多元线性回归；神经网络

Abstract: With the rise of self-drive travel, the market of China's recreational vehicle (RV) is showing a vigorous development. The purpose of this paper is to explore the main driving factors behind the sales market of RV, forecast and analyze the changing trend of the sales of RV. First of all, based on the relevant statistical data in the past ten years, through the Pearson correlation coefficient calculation, the main driving factors affecting the sales market of RV are per capita gross domestic product, per capita disposable income, highway miles and the number of people over 60 years old. Then, using the methods of time series analysis, multiple linear regression and neural network, it constructs the sales model of RVs, and forecast the sales volume of RVs in China in the next three years. The predicted sales volume is compared with the actual sales volume to determine the prediction accuracy of each model. The results show that the highest goodness of fit is the multiple linear regression model, which predicts the sales volume of China's passenger cars in 2024－2026 are 15 737, 17 215 and 18 693 respectively. The research results can provide an important basis for making production plan, adjusting market strategy and optimizing resource allocation.

Key words: recreational vehicle; sales forecast; time series; multiple linear regression; neural network