本文關鍵詞： 滬深300股指期貨 分形市場假說 有效市場假說　出處：《南京大學》2017年碩士論文　論文類型：學位論文

【摘要】：作為現(xiàn)代金融研究的理論基礎,有效市場假說的假設條件與實際情況并不完全相符,且無法解釋許多現(xiàn)實存在的現(xiàn)象。資本市場的本質(zhì)是非線性復雜系統(tǒng),而在非線性理論之中,分形理論具有廣泛適應性與實用性。它在金融領域形成的具體應用被稱之為分形市場假說(Fractal Market Hypothesis,FMIH)。對于中國證券市場而言,滬深300股指期貨的正式運營代表著一個全新金融工具的出現(xiàn),可以期待它在將的來的某一天充分發(fā)揮成熟的股指期貨市場所具備的套期保值、價格發(fā)現(xiàn)、風險規(guī)避三大功能。本文借助分形市場假說相關理論、方法與工具對滬深300股指期貨市場進行研究。首先闡述了有效市場假說與分形市場假說的起源、發(fā)展、理論模型及適用范圍,并將兩者進行對比,認為分形市場理彌補了有效市場假說脫離真實市場的缺陷、是有效市場假說的拓展、為金融市場研究開拓了新的方向。隨后通過描述性檢驗、J-B檢驗和QQ圖、L-B檢驗和BDS檢驗、STABLE分布參數(shù)對比、R/S分析等方法,說明該市場中存在非線性、自相似性、長記憶性的分形特征�？梢哉J為滬深300股指期貨市場是一個分形市場,其循環(huán)周期約為198天。本文還通過改進盒計數(shù)法、MF-DFA方法對樣本的整體特征進行度量,證明了滬深300股指期貨市場上存在多重分形結(jié)構(gòu),小幅波動具有(更強的)持久性特征,大幅波動具有(更強的)反持久性特征。然后運用控制變量法與MF-DFA方法結(jié)合,探究了滬深300股指期貨市場多重分形結(jié)構(gòu)的成因,結(jié)論是長記憶性、厚尾性和極端波動均是該市場存在多重分形特征的成因,其中厚尾性起的作用最大。最后推測了該市場存在分形結(jié)構(gòu)的具體原因。
[Abstract]:As the theoretical basis of modern financial research, the hypothesis conditions of efficient market hypothesis are not completely consistent with the actual situation, and can not explain many phenomena existing in reality. The essence of capital market is a nonlinear complex system. In the nonlinear theory, fractal theory has wide adaptability and practicability. Its specific application in the financial field is called fractal Market hypothesis. The official operation of Shanghai and Shenzhen 300 stock index futures represents the emergence of a new financial instrument. It can be expected to give full play to the hedging and price discovery in the mature stock index futures market one day. With the help of the theory of fractal market hypothesis, this paper studies the stock index futures market of Shanghai and Shenzhen 300 by means of methods and tools. First of all, the origin and development of efficient market hypothesis and fractal market hypothesis are expounded. The theoretical model and the scope of application are compared, and it is considered that the fractal market theory makes up the defect of the efficient market hypothesis from the real market, and it is an extension of the efficient market hypothesis. This paper opens up a new direction for the study of financial market. Then, by means of descriptive test J-B test, QQ diagram L-B test and BDS test comparing the distribution parameters of STABLE with R / S analysis, it is shown that there are nonlinear and self-similarity in this market. It can be considered that the Shanghai and Shenzhen 300 stock index futures market is a fractal market with a cycle period of about 198 days. This paper also measures the overall characteristics of the samples by the improved box counting method and MF-DFA method. It is proved that there is multifractal structure in Shanghai and Shenzhen 300 stock index futures market. Small volatility has (stronger) persistence feature, and large volatility has (stronger) anti-persistence feature. Then the control variable method is combined with MF-DFA method. This paper probes into the causes of multifractal structure of Shanghai and Shenzhen 300 stock index futures market. The conclusion is that long memory, thick tail and extreme fluctuation are the causes of multifractal characteristics in this market. The effect of thick tail is the most important. Finally, the specific reason of fractal structure in this market is inferred.
【學位授予單位】：南京大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：F224;F724.5
，

本文編號：1508106