本项目研究区间计量经济模型及其在经济分析和金融市场的应用。传统计量模型对点样本建模，但区间样本比点样本含有更丰富的信息。区间计量模型建立直接对区间样本建模、对区间总体进行统计推断的计量理论体系；基于区间样本较点样本的信息优势，区间计量模型将提高传统计量建模技术的统计推断效率和预测精度，推动计量经济理论的发展。区间计量模型可应用于测算金融市场风险、预测重要经济变量，为政府和相关行业的企业提供分析工具和决策支持。具体研究内容包括：（1）建立区间计量模型体系，包括区间时间序列、区间横截面以及区间面板变量等三个计量模型子体系，并分别建立线性、非线性和联立方程等多种模型形式（2）提出区间计量模型的多种参数估计方法（3）建立区间计量模型参数的多种假设检验方法并证明估计量的渐进有效（4）应用区间计量模型对我国外资外贸、国际大宗商品期货价格、外汇市场、全球主要股指等经济分析与金融市场问题进行建模与预测。

Econometrics has been concerned with point-valued data modeling. The proposed research is a first attempt to model the dynamics of interval-valued data. There is a large body of interval-valued observations in economics and finance, which often contain richer information than point-based observations since an interval number captures both 'range' and 'level' characteristics of the underlying probability space. Also, interval forecasts maybe of direct interest in practice. The informational advantage of interval data can be exploited for more efficient econometric estimation and inference...To forecast intervals and to explore the potential gain of using interval data over using point-valued data, the proposed research will develop a novel methodology of interval modeling in econometrics, for interval-valued time series data, interval-valued cross-sectional data and interval-valued panel data, by which one can use the observed interval data to infer the population of random intervals, and to use it for forecasts and other applications in economics and finance. As a counterpart of traditional point-based econometrics, the novel theory includes a new class of conditional interval models corresponding to the three categories of interval data aforementioned, the corresponding asymptotic theories for estimation, testing and inference. Interval econometric models will be applied in modeling and forecasting economic fundamentals as well as high-frequency data in financial markets, such as foreign trade, global bulk prices, foreign exchange rates, stock market indices and so on.