發(fā)布時(shí)間：2018-09-07 07:30

【摘要】：本文選用人工股指期貨平臺(tái)自帶的j30指數(shù)作為現(xiàn)貨指數(shù)序列,考察在人工股指期貨市場的特定市場結(jié)構(gòu)中,不同投資策略的投資者進(jìn)行下單交易后產(chǎn)生的市場價(jià)格序列、成交量等涌現(xiàn)出來的宏觀特性,并運(yùn)用描述性統(tǒng)計(jì)量,ARMA模型和ARCH-LM檢驗(yàn)來度量收益率的尖峰厚尾和波動(dòng)聚集特征。人工股指期貨市場(U-Mart)的市場架構(gòu)為股指期貨的單市場交易,我們只能在其中買賣期貨合約,不存在期現(xiàn)市場的雙向作用機(jī)制,也不存在跨市場套利行為。因此,現(xiàn)貨指數(shù)序列是取自歷史的真實(shí)數(shù)據(jù)或者隨機(jī)生成的,是外生于期貨市場的。我們采用人工期貨市場的競價(jià)機(jī)制是集合競價(jià),并采用固定算法規(guī)則撮合,無漲停和跌停的限制,也無摩擦和手續(xù)費(fèi)用,并且允許賣空,其交易條件是相當(dāng)寬松的。這樣一來我們就可以在一定額度的貸款下來更好地實(shí)現(xiàn)和觀測操作策略對(duì)于市場宏觀特性的影響。 經(jīng)過實(shí)驗(yàn)研究結(jié)論證明:在實(shí)驗(yàn)一中,隨機(jī)交易生成的股指期貨價(jià)格非常平穩(wěn),僅在1980-2100間小范圍小幅度波動(dòng),波動(dòng)性低,成交量也很小,流動(dòng)性較差,從收益率特性來看,實(shí)驗(yàn)一服從了收益率的正態(tài)分布假設(shè),并沒有出現(xiàn)現(xiàn)實(shí)金融市場中出現(xiàn)的收益率的尖峰厚尾性,同時(shí)經(jīng)過拉格朗日乘數(shù)檢驗(yàn)后,也沒有發(fā)現(xiàn)波動(dòng)聚集性。在實(shí)驗(yàn)二中,移動(dòng)平均投資者的加入使得期貨價(jià)格波動(dòng)性略微提高,在2020-2200間波動(dòng),成交量也略微提高,流動(dòng)性增加。從收益率曲線的特性來看,移動(dòng)平均投資者的加入使得曲線開始偏離正態(tài)分布,盡管這種偏離程度不高,但是我們?nèi)匀豢梢钥吹郊夥搴裎蔡卣鞯某霈F(xiàn),同時(shí)我們也發(fā)現(xiàn)了價(jià)格收益率的波動(dòng)群集特征。在實(shí)驗(yàn)三中,動(dòng)量投資者的加入使得股指期貨價(jià)格開始劇烈波動(dòng),成交量也大幅增加了。收益率曲線開始明顯的偏離正態(tài)分布,形成典型的尖峰厚尾特征,同時(shí)波動(dòng)集群特征也明顯出現(xiàn),由此我們可以得出趨勢投資者是造成金融市場中尖峰厚尾和波動(dòng)群集特性的重要原因。
[Abstract]:This paper selects the J30 index of artificial stock index futures platform as the spot index sequence to investigate the market price sequence of investors with different investment strategies in the specific market structure of artificial stock index futures market. The macroscopical characteristics of turnover and so on are presented. The ARMA model of descriptive statistics and ARCH-LM test are used to measure the characteristics of peak, thick tail and volatility aggregation of yield. The market structure of artificial stock index futures market (U-Mart) is the single market trading of stock index futures, in which we can only buy and sell futures contracts, there is no two-way mechanism of the current market, and there is no cross-market arbitrage behavior. Therefore, the spot index series is derived from historical real data or randomly generated, is born out of the futures market. In the artificial futures market, the bidding mechanism is set bidding, and the fixed algorithm is used to set the price. There is no limit of the limit and limit, no friction and formalities, and the short selling is allowed. The trading conditions are quite loose. In this way, we can better realize and observe the effect of the operation strategy on the macro characteristics of the market with a certain amount of loan. The experimental results show that in the first experiment, the price of stock index futures generated by random trading is very stable, and only fluctuates in a small range between 1980 and 2100. The volatility is low, the turnover is very small, and the liquidity is poor. In the first experiment, the hypothesis of normal distribution of rate of return is given, and there is no sharp and thick tail of yield in real financial market. At the same time, after the Lagrange multiplier test, there is no volatility aggregation. In experiment 2, the addition of moving average investors slightly increased futures price volatility, fluctuating between 2020-2200, trading volume and liquidity. From the point of view of the characteristics of the yield curve, the addition of moving average investors makes the curve deviate from the normal distribution. Although the deviation is not high, we can still see the appearance of the peak and thick tail. At the same time, we also found the volatility cluster characteristics of price return. In experiment 3, momentum investor's addition made stock index futures price fluctuate sharply and turnover increased. The yield curve began to deviate from the normal distribution obviously, forming the typical peak and thick tail characteristics, and the characteristic of fluctuation cluster also appeared obviously. From this we can conclude that trend investors are the important reasons for the characteristics of the peak and the volatility cluster in the financial market.
【學(xué)位授予單位】：天津大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2012
【分類號(hào)】：F224;F830.91

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 馬進(jìn)勝;邱菀華;;基于主體的計(jì)算金融學(xué)綜述[J];北京航空航天大學(xué)學(xué)報(bào)(社會(huì)科學(xué)版);2007年02期

2 邢麗卿;孫紹榮;;基于Agent的計(jì)算實(shí)驗(yàn)金融學(xué):原理、方法和挑戰(zhàn)[J];廣西師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2008年01期

3 張強(qiáng);楊淑娥;戴耀華;;中國股市動(dòng)量策略和反轉(zhuǎn)策略的實(shí)證分析[J];華東經(jīng)濟(jì)管理;2007年05期

4 張維;趙帥特;熊熊;張永杰;;計(jì)算實(shí)驗(yàn)金融、技術(shù)規(guī)則與時(shí)間序列收益可預(yù)測性[J];管理科學(xué);2008年03期

5 王永宏,趙學(xué)軍;中國股市“慣性策略”和“反轉(zhuǎn)策略”的實(shí)證分析[J];經(jīng)濟(jì)研究;2001年06期

6 張人驥,朱平方,王懷芳;上海證券市場過度反應(yīng)的實(shí)證檢驗(yàn)[J];經(jīng)濟(jì)研究;1998年05期

7 高志杰;;中美小麥期貨價(jià)格波動(dòng)特征比較[J];中國農(nóng)業(yè)科技導(dǎo)報(bào);2006年01期

8 周琳杰;中國股票市場動(dòng)量策略贏利性研究[J];世界經(jīng)濟(jì);2002年08期

9 陳工孟,芮萌;中國股票市場的股票收益與波動(dòng)關(guān)系研究[J];系統(tǒng)工程理論與實(shí)踐;2003年10期

相關(guān)博士學(xué)位論文 前1條

1 張永杰;信念、策略、資產(chǎn)價(jià)格與投資收益[D];天津大學(xué);2007年

相關(guān)碩士學(xué)位論文 前1條

1 李蓉蓉;股票市場投資策略仿真研究[D];天津大學(xué);2009年

，

本文編號(hào)：2227615