Private Car Ownership and City Development: A Method Combining Machine Learning and Time Series

The traffic development level of a city can often reflect the economic development level of the city. Therefore, the introduction of car ownership to predict the level of urban economic development is one of the necessary means to reflect the relationship between the two parameters. GDP is an important indicator to measure economic development. If it can be revealed the law of GDP change and make a high-precision prediction of short-term GDP, it is of great practical significance to the overall planning of the macro economy. In this paper, the data of population density, a permanent population, fixed investment, and private car ownership are predicted by the time series model in the next five years. Thus, neural networks and conventional regression has been used to forecast GDP and compare. It is of great significance to explore the method of future GDP forecast.