中文摘要：

企业将资产运用于生产经营活动,并由此赚取更多的资产,即产生公司的收入.因此企业资产与收入之间必定存在一定的相关关系.在对上市公司总资产与营业收入进行一般线性拟合的基础之上,采用分位回归模型对上市公司的总资产与营业收入的关系进行深入剖析.结果表明,传统的线性模型只能揭示出总资产与营业收入呈正相关关系,而分位回归方法能更好地看出,高分位点营业收入的企业在提高一定总资产时,会更能促进营业收入的增长.由于收集到的数据中存在离群点,在第5节讨论了线性分位回归模型的统计诊断,类比于一般线性模型的R square得到不同分位点上的R square.通过删除离群点的处理,得出分位回归模型比一般线性模型更加稳健,数据在高分位点的拟合效果更好一些.

英文摘要：

Assets are used in the production and business operation activities by enterprises,and thus enterprises will make more assets.That is to say,it can generate the enterprises＇ revenue.So,there must be a certain correlation between enterprise＇ revenue and assets.Based on the general linear fitting of the listed companies＇ total assets and revenue,this paper adopt quantile regression model to make an in- depth analysis of the relationship between total assets and revenue.It turn out that,there is a positive correlation between total assets and revenue.When enterprises improve a certain assets,the enterprises which have high quantile in revenu will make greater increase in revenue.Due to the outliers in the data collected,the paper in section 5 discusses the problem about statistical diagnosis in linear quantile regression model,in analogy to the R square in general linear model,we get the R square in different quantiles.By removing the outliers,draw the quantile regression model is more robust than general linear model.As to the data in this paper,the fitting effect in high quantile is better.