本文選題：限價(jià)指令簿　切入點(diǎn)：動(dòng)態(tài)建模　出處：《廈門(mén)大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：我國(guó)A股市場(chǎng)是典型的指令驅(qū)動(dòng)市場(chǎng),與傳統(tǒng)報(bào)價(jià)驅(qū)動(dòng)方式不同,投資者可以提交市價(jià)單和限價(jià)單,這些訂單按照“價(jià)格優(yōu)先,時(shí)間優(yōu)先”的原則自動(dòng)匹配成交,而未成交的限價(jià)單累積形成限價(jià)指令簿。由于限價(jià)指令簿中含有大量偏離最優(yōu)報(bào)價(jià)的訂單,很多學(xué)者認(rèn)為這些未成交的訂單具有一定的信息含量,從限價(jià)指令簿中提取的信息能夠極大地幫助股價(jià)短期預(yù)測(cè)。除此之外,在具體交易細(xì)節(jié)中,投資者還面臨訂單選擇的決策問(wèn)題,這也取決于限價(jià)指令簿的狀態(tài)和變化特征。因此,建立模型來(lái)刻畫(huà)限價(jià)指令簿信息顯得十分必要,而本文的目的就是試圖建立限價(jià)指令簿的動(dòng)態(tài)模型。 本文建立了動(dòng)態(tài)限價(jià)指令簿模型。首先,由于限價(jià)指令簿的動(dòng)態(tài)變化是各類訂單流累積的表現(xiàn),在完成限價(jià)指令簿靜態(tài)描述的基礎(chǔ)上,本文分析了訂單流信息與動(dòng)態(tài)限價(jià)指令簿狀態(tài)變化的聯(lián)系。根據(jù)訂單流的特征,本文設(shè)定了訂單流的具體隨機(jī)過(guò)程,得到了動(dòng)態(tài)限價(jià)指令簿的三因素(買(mǎi)賣(mài)價(jià)差、中間價(jià)格和買(mǎi)賣(mài)壓力)模型。在一定假設(shè)下將無(wú)限維的指令簿動(dòng)態(tài)過(guò)程轉(zhuǎn)換為最優(yōu)報(bào)價(jià)與最優(yōu)買(mǎi)賣(mài)量的動(dòng)態(tài)過(guò)程。上證50成分股指令簿數(shù)據(jù)的實(shí)證表明,模型很好地還原了市場(chǎng)特征,并且得到了短期內(nèi)指令簿信息對(duì)股價(jià)的影響遠(yuǎn)高于市場(chǎng)外部信息的重要結(jié)論。 為了分析動(dòng)態(tài)限價(jià)指令簿模型是否能夠幫助下單決策,我們對(duì)比分析了訂單執(zhí)行等待時(shí)間的經(jīng)驗(yàn)分布,以及三因素模型與兩因素(不含指令簿信息)模型下的理論分布。實(shí)證結(jié)果表明,當(dāng)市場(chǎng)初始狀態(tài)為賣(mài)方壓力時(shí),加入指令簿信息能夠幫助我們更好估計(jì)限價(jià)賣(mài)單不被執(zhí)行的風(fēng)險(xiǎn)；而市場(chǎng)為買(mǎi)方壓力時(shí),加入指令簿信息卻放大了估計(jì)誤差。本文認(rèn)為這是因?yàn)橘I(mǎi)賣(mài)壓力對(duì)訂單執(zhí)行等待時(shí)間的影響是非對(duì)稱的。經(jīng)驗(yàn)分布中,買(mǎi)方壓力將顯著提高限價(jià)賣(mài)單被執(zhí)行的概率和降低限價(jià)買(mǎi)單被執(zhí)行的概率,且壓力越大影響越明顯；而賣(mài)方壓力對(duì)限價(jià)賣(mài)單執(zhí)行概率的影響并不隨壓力大小線性增加,這說(shuō)明賣(mài)方壓力逐漸釋放過(guò)程中存在買(mǎi)賣(mài)雙方激進(jìn)程度的輪換,進(jìn)一步揭示買(mǎi)賣(mài)雙方耐心程度的差別。而模型中買(mǎi)賣(mài)壓力對(duì)價(jià)格影響對(duì)稱的假定偏離了實(shí)際情況,存在進(jìn)一步改進(jìn)的空間。
[Abstract]:China's A-share market is a typical order-driven market. Unlike the traditional pricing drive, investors can submit market price orders and price limit orders, and these orders can automatically match and conclude in accordance with the principle of "price first, time first". Because the price limit order book contains a large number of orders that deviate from the optimal quotation, many scholars believe that these unsold orders have a certain amount of information. The information extracted from the price limit book can greatly help short-term stock price forecasts. In addition to the details of the transaction, investors are also faced with the decision to choose an order. Therefore, it is necessary to establish a model to depict the information of the price limit instruction book, and the purpose of this paper is to try to establish the dynamic model of the price limit instruction book. In this paper, the dynamic pricing instruction book model is established. Firstly, because the dynamic change of the price limit instruction book is the performance of all kinds of order flow accumulation, on the basis of completing the static description of the price limit instruction book, This paper analyzes the relationship between the information of order flow and the change of state of dynamic limit price instruction book. According to the characteristics of order flow, this paper sets up the specific stochastic process of order flow, and obtains three factors of dynamic price limit instruction book (the difference of buying and selling price). Under certain assumptions, the infinite dimensional dynamic process of instruction book is transformed into the dynamic process of optimal quotation and optimal amount of trading. The model restores the market characteristics well, and obtains the important conclusion that the short term instruction book information has much higher influence on the stock price than the external market information. In order to analyze whether dynamic price limit instruction book model can help order decision, we compare and analyze the empirical distribution of waiting time for order execution. The empirical results show that when the initial state of the market is under the pressure of the seller, the theoretical distribution of the three factors model and the two-factor model (excluding instruction book information) is obtained. Adding instruction book information can help us better estimate the risk that the price limit order will not be executed; and when the market is under pressure from the buyer, This paper argues that this is because the influence of buying and selling pressure on the waiting time for order execution is asymmetric. Buyer pressure will significantly increase the probability of price limit orders being executed and reduce the probability of price limit purchase being executed, and the greater the pressure is, the more obvious the effect will be; however, the influence of seller pressure on the price limit order execution probability does not increase linearly with the pressure. This indicates that there is a radical rotation between the seller and the seller during the gradual release of the seller's pressure, which further reveals the difference in the degree of patience between the seller and the seller, while the assumption that the seller's pressure has a symmetrical effect on the price deviates from the actual situation. There is room for further improvement.
【學(xué)位授予單位】：廈門(mén)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：F224;F832.51

【參考文獻(xiàn)】

相關(guān)期刊論文 前4條

1 屈文洲,吳世農(nóng);中國(guó)股票市場(chǎng)微觀結(jié)構(gòu)的特征分析——買(mǎi)賣(mài)報(bào)價(jià)價(jià)差模式及影響因素的實(shí)證研究[J];經(jīng)濟(jì)研究;2002年01期

2 鄭振龍;戴嵩;;買(mǎi)賣(mài)價(jià)差與限價(jià)指令簿信息:基于時(shí)變MRR模型的實(shí)證研究[J];金融評(píng)論;2011年05期

3 楊之曙,李子奈;上海股市日內(nèi)流動(dòng)性——深度變化實(shí)證研究[J];金融研究;2003年06期

4 陳淼鑫;;限價(jià)指令簿與指令驅(qū)動(dòng)市場(chǎng)研究述評(píng)[J];廈門(mén)大學(xué)學(xué)報(bào)(哲學(xué)社會(huì)科學(xué)版);2011年06期

，

本文編號(hào)：1646493