金融区间序列分析及其预测的初步评价

发布时间：2018-07-11 16:40

本文选题：金融区间序列 + 模糊线性规划　；参考：《数理统计与管理》2010年01期

【摘要】：金融市场中的实际观测数据,除随机性外往往还带有模糊性,这样的观测数据通常以观测区间的形式给出,例如,当我们谈及某日的上证指数时,其观测值总是在最低点与最高点之间波动。观测值的这种不确定性来自于多重隶属现象,而非随机现象,我们称这种不确定性为模糊性。不确定性问题通常含有两种意义上的分类:一类是随机不确定性,人们依靠概率统计方法进行处理;另一类是非随机性的不确定性,即为模糊性问题,通常利用模糊集合理论来进行研究。本文将在模糊数学理论基础上,利用回归分析方法,构建模糊金融时间序列模型,并利用FLP(模糊线性规划)方法来估计模型的未知参数。为了合理评价拟合效果,我们将根据测量模糊集合间的择近原则,给出利用样本平均贴近度来评价模型拟合效果的一个准则。实证研究将讨论金融模糊时间观测序列的建模、参数估计,并对模型的实际拟合效果和预测效果做初步评价。
[Abstract]:The actual observations in financial markets tend to be fuzzy in addition to randomness, and such observations are usually given in the form of observation intervals, for example, when we talk about the Shanghai Stock Exchange index on a certain day. Its observations always fluctuate between the lowest and highest points. The uncertainty of observations comes from multiple membership rather than random phenomena, which we call fuzziness. Uncertainty problems usually contain two kinds of classification: one is random uncertainty, people rely on probability and statistics method to deal with it, the other is non-random uncertainty, that is, fuzzy problem. The fuzzy set theory is usually used to study. In this paper, based on the theory of fuzzy mathematics, the fuzzy financial time series model is constructed by means of regression analysis, and the unknown parameters of the model are estimated by using FLP (fuzzy linear programming) method. In order to evaluate the fitting effect reasonably, we will give a criterion to evaluate the model fitting effect by using the average closeness degree of the sample according to the principle of selecting the near between the measured fuzzy sets. The empirical study will discuss the modeling and parameter estimation of the financial fuzzy time series, and make a preliminary evaluation on the actual fitting effect and prediction effect of the model.
【作者单位】：四川大学数学学院;上海财经大学会计学院;
【基金】：国家社会科学基金统计学项目(07BTJ003)
【分类号】：F830;F224

【相似文献】