中文摘要：

本文基于分位数的回归理论与方法，提出了一个新的经济计量模型：分位数局部调整模型，并给出了其数学表示、参数估计与预测方法等一整套建模技术。分位数局部调整模型能够细致地给出响应变量在各个分位点上的条件分位数，便于揭示响应变量位置、散布与形状等动态调整过程的全景信息，从而得到比均值局部调整模型更为深刻的结果。最后，将分位数局部调整模型应用于中国货币需求分析，结果显示，在货币需求的不同阶段，不仅调整速度不同，调整方式也呈现出非对称性；M1存在货币失踪之谜现象，而M2却在条件密度第一个最优区域实现了供求均衡；最优货币需求条件密度曲线较为分散，这为央行制定货币政策预留了足够的空间。

英文摘要：

A new econometric model, named quantile partial adjustment model, is proposed in the paper based on theory and methodology of quantile regression. A set of modeling methodology, includes mathematics expression, parameter estimation and forecasting, is studied in detail. The new model can be used to ob- tain conditional quantiles, which is very useful for describing location, scale and shape of dynamic adjustment process. Finally, the new model is applied to analyze Chinese money demand. Empirical results show that the speed of adjustment is varying with different stage of money demand, and the pattern of adjustment is asymmetry. Mystery of missing money exists in M1. However, supply and demand balance for M2 is obtained in the first optimal area of conditional density. The disperse of conditional density of optimal money demand provide a large space for monetary policy makers.