Sierpinski格点渗流股票价格模型及预测分析

发布时间：2018-02-13 11:24

本文关键词： 渗流理论 Sierpinski格点地毯股票价格模型预测分析复杂度多重分形性　出处：《北京交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：金融物理学的诞生是金融研究史上的又一次革命,它通过物理统计方法帮助人们加深对金融证券市场的了解,受到学者的广泛关注。本文旨在构建金融物理价格模型,并利用该模型进行预测分析。本文将统计物理模型中的渗流模型应用到金融建模中,构建了位于分形地毯上的股票价格波动模型。通过对真实市场数据和预测数据进行统计性质和复杂性质的分析,探究了金融市场相关特性,并验证了模型的合理性和有效性,同时对构造的股票价格模型进行实证分析,提供相关预测指导。本文共五个章节,内容如下:第一章介绍金融物理学研究背景,从金融研究者和金融市场参与者的角度论述了本文的研究意义与价值,并详尽阐述学者相关研究,对本文的模型构建、统计量选取、统计分析方法应用这三个方面的创新内容进行展示。第二章介绍了各类常用预测方法并进行评析,讨论了经济预测中可能存在的问题。第三章完成具体价格模型的构建,并对模型原理及其中参数进行相关讨论,探究不同参数对模型统计特性的影响。第四章对模型预测数据和真实市场数据进行相关对比分析:第一种为复杂度(LZC)分析,定量计算了金融市场的复杂程度,既对全球各主要金融市场复杂程度进行了分析,也对比出我国金融市场和预测数据复杂度的相似性,并通过拟合程度确定模型中的待定参数;第二种为多重分形性(MF-DCCA)分析,研究了我国金融市场和预测数据的多重分形性,通过对多指标的计算和比对,证明了真实市场和预测数据均存在多重分形性,且指标数值相近。上述模拟预测分析从金融市场内在特性论证了金融价格模型的有效性。考虑到研究的实用价值,本章对其进行样本外实证检验,效果较好,并根据目前经济状况给出30个预测值,但实际预测还需使用者结合市场即时情况和自身判断,对模型中初始值等参数进行设置与调节,再进行预测。第五章对上文研究内容进行总结。
[Abstract]:The birth of financial physics is another revolution in the history of financial research. It helps people to understand the financial securities market more deeply through the method of physical statistics. In this paper, the seepage model in the statistical physical model is applied to the financial modeling. The stock price volatility model on the fractal carpet is constructed. By analyzing the statistical and complex properties of the real market data and the forecast data, the related characteristics of the financial market are explored, and the rationality and validity of the model are verified. At the same time, the empirical analysis of the stock price model is carried out to provide guidance for prediction. There are five chapters in this paper. The contents are as follows: chapter 1 introduces the background of financial physics research. From the perspective of financial researchers and financial market participants, this paper discusses the significance and value of this study, and elaborates the relevant research of scholars, the model construction, the selection of statistics, The second chapter introduces various kinds of commonly used forecasting methods, discusses the possible problems in economic forecasting, and completes the construction of specific price model in chapter three. The model principle and its parameters are discussed, and the influence of different parameters on the statistical characteristics of the model is discussed. Chapter 4th compares the model prediction data with the real market data: the first is the complexity analysis. The complexity degree of the financial market is calculated quantitatively. The complexity of the major financial markets is analyzed, and the similarity of the complexity between the financial market and the forecast data in China is compared, and the undetermined parameters in the model are determined by the degree of fitting. The second is multifractal analysis, which studies the multifractal of financial market and forecast data in China. Through the calculation and comparison of multiple indexes, it is proved that there are multifractal in real market and forecast data. And the numerical value of the index is similar. The simulation and forecast analysis prove the validity of the financial price model from the intrinsic characteristics of the financial market. Considering the practical value of the research, this chapter carries out the empirical test on it outside the sample, and the effect is good. According to the current economic situation, 30 prediction values are given, but the actual prediction still needs the users to set and adjust the initial value of the model in combination with the market situation and their own judgment. Chapter 5th summarizes the contents of the above research.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F832.51

【参考文献】