【摘要】：近年來(lái),國(guó)內(nèi)外上市公司舞弊丑聞層出不窮,給投資者帶來(lái)巨大的投資風(fēng)險(xiǎn)和傷害的同時(shí),沉重打擊了社會(huì)公眾對(duì)會(huì)計(jì)界和資本市場(chǎng)的信心。因此,如何有效識(shí)別企業(yè)舞弊行為成為會(huì)計(jì)理論界、實(shí)務(wù)界以及監(jiān)管部門關(guān)注的重中之重。實(shí)證研究表明,模型舞弊識(shí)別效果優(yōu)于舞弊案例分析,而有效的舞弊風(fēng)險(xiǎn)識(shí)別模型的構(gòu)建離不開完善的舞弊識(shí)別指標(biāo)和恰當(dāng)?shù)淖R(shí)別方法。目前,在舞弊識(shí)別指標(biāo)方面的研究已經(jīng)比較完善,但舞弊識(shí)別模型方面的研究較少。隨著人工智能技術(shù)的不斷發(fā)展和廣泛應(yīng)用,人工神經(jīng)網(wǎng)絡(luò)技術(shù)開始應(yīng)用于舞弊識(shí)別領(lǐng)域。其中,以BP和LVQ神經(jīng)網(wǎng)絡(luò)在舞弊識(shí)別領(lǐng)域的應(yīng)用最為廣泛,舞弊識(shí)別率較高。本文在此背景下深入探究這兩種神經(jīng)網(wǎng)絡(luò)技術(shù),用同一舞弊樣本檢驗(yàn)這兩種模型的舞弊識(shí)別率,并在此基礎(chǔ)上提出優(yōu)化的基于BP-LVQ的組合神經(jīng)網(wǎng)絡(luò)舞弊風(fēng)險(xiǎn)識(shí)別模型。本文查閱整理國(guó)內(nèi)外相關(guān)文獻(xiàn)后,在第二章文獻(xiàn)綜述部分簡(jiǎn)單闡述了國(guó)際上最為流行的六種管理舞弊動(dòng)機(jī)與成因理論以及國(guó)內(nèi)流行的舞弊動(dòng)機(jī)與成因觀點(diǎn),梳理歸納了舞弊風(fēng)險(xiǎn)識(shí)別指標(biāo)和舞弊風(fēng)險(xiǎn)識(shí)別方法的相關(guān)國(guó)內(nèi)外文獻(xiàn)資料,明確舞弊風(fēng)險(xiǎn)識(shí)別模型的研究現(xiàn)狀、研究成果和現(xiàn)有的不足之處。在此基礎(chǔ)上提出本文選用BP、LVQ神經(jīng)網(wǎng)絡(luò)技術(shù)作為舞弊風(fēng)險(xiǎn)識(shí)別模型的理由,并在第三章詳細(xì)介紹了人工神經(jīng)網(wǎng)絡(luò)技術(shù)的特點(diǎn)與分類、BP神經(jīng)網(wǎng)絡(luò)和LVQ神經(jīng)網(wǎng)絡(luò)的結(jié)構(gòu)及運(yùn)作機(jī)制。第四章主要是樣本選取和舞弊風(fēng)險(xiǎn)識(shí)別指標(biāo)篩選。本文選取2010年到2014年發(fā)生舞弊的506家上市公司作為舞弊樣本,按照Beasley原則一比一確定非舞弊的配對(duì)樣本公司506家,以此作為研究樣本。將根據(jù)文獻(xiàn)梳理出的識(shí)別效果較好的舞弊風(fēng)險(xiǎn)識(shí)別指標(biāo)作為最初的指標(biāo)體系,通過(guò)配對(duì)樣本T檢驗(yàn)以及主成分分析消除共線性問(wèn)題后,最終刪選出識(shí)別效果最好的10個(gè)指標(biāo)。第五章主要對(duì)BP、LVQ神經(jīng)網(wǎng)絡(luò)模型的舞弊風(fēng)險(xiǎn)識(shí)別效果進(jìn)行檢驗(yàn),并對(duì)兩種識(shí)別模型的舞弊判別效果進(jìn)行分析。第六章在分析BP、LVQ神經(jīng)網(wǎng)絡(luò)模型各自優(yōu)缺點(diǎn)的基礎(chǔ)上提出構(gòu)建基于BP-LVQ的組合神經(jīng)網(wǎng)絡(luò)模型,介紹了組合模型的構(gòu)建原理和思路,用同一研究樣本檢驗(yàn)組合模型的舞弊識(shí)別效果,發(fā)現(xiàn)組合模型的舞弊判別率顯著高于單個(gè)神經(jīng)網(wǎng)絡(luò)模型,用2015年舞弊樣本數(shù)據(jù)進(jìn)行穩(wěn)定性檢驗(yàn)后發(fā)現(xiàn),組合模型的舞弊識(shí)別效果更好而且判別率穩(wěn)定。最后,根據(jù)上文理論分析和實(shí)證研究總結(jié)全文,分析本次研究中的不足之處,并對(duì)未來(lái)舞弊風(fēng)險(xiǎn)識(shí)別模型的發(fā)展提出展望。
[Abstract]:In recent years, fraud scandals of listed companies at home and abroad have emerged one after another, which bring huge investment risks and injuries to investors, and at the same time, have dealt a heavy blow to the public's confidence in accounting and capital markets. Therefore, how to effectively identify corporate fraud has become the most important concern of accounting theory, practice and regulatory authorities. Empirical research shows that the effectiveness of fraud identification model is better than that of fraud case analysis, and the effective fraud risk identification model can not be constructed without perfect fraud identification index and appropriate identification method. At present, the research on fraud identification index has been perfect, but the research on fraud identification model is less. With the development and wide application of artificial intelligence technology, artificial neural network technology has been applied to the field of fraud recognition. Among them, BP and LVQ neural networks are the most widely used in the field of fraud identification, fraud recognition rate is high. In this context, this paper explores the two neural network technologies, and uses the same fraud sample to test the fraud recognition rate of the two models, and on this basis, an optimized combined neural network fraud risk identification model based on BP-LVQ is proposed. In the second chapter of literature review, the author briefly expounds the six most popular theories of management fraud motivation and cause, and the domestic popular theories of fraud motivation and cause of formation. Combing and summarizing the relevant domestic and foreign literature data of fraud risk identification index and fraud risk identification method, clarifying the research status of fraud risk identification model, research results and existing deficiencies. On this basis, the reason for choosing BP,LVQ neural network technology as the fraud risk identification model is put forward. In chapter 3, the characteristics and classification of artificial neural network technology are introduced in detail. The structure and operation mechanism of BP neural network and LVQ neural network. The fourth chapter is mainly about sample selection and fraud risk identification index selection. In this paper, 506 listed companies with fraud from 2010 to 2014 are selected as fraud samples, and 506 non-fraudulent matched sample companies are determined according to Beasley principle. In this paper, the fraud risk identification index, which has good identification effect according to the literature, is taken as the initial index system. After eliminating the collinearity problem by paired sample T test and principal component analysis, the best 10 indexes are selected. Chapter 5 mainly tests the effect of fraud risk identification of BP,LVQ neural network model, and analyzes the fraud discrimination effect of two kinds of identification models. In chapter 6, on the basis of analyzing the advantages and disadvantages of BP,LVQ neural network model, the author proposes to construct the combined neural network model based on BP-LVQ, and introduces the construction principle and train of thought of the combined neural network model. Using the same sample to test the fraud identification effect of the combined model, it is found that the fraud discrimination rate of the combined model is significantly higher than that of the single neural network model. The combined model has better effect of fraud identification and stable discriminant rate. Finally, according to the theoretical analysis and empirical research summarized the full text, analyzes the shortcomings of this study, and puts forward the future development of fraud risk identification model.
【學(xué)位授予單位】：杭州電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：F275;F832.51

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李揚(yáng);李竟翔;馬雙鴿;;不平衡數(shù)據(jù)的企業(yè)財(cái)務(wù)預(yù)警模型研究[J];數(shù)理統(tǒng)計(jì)與管理;2016年05期

2 宋曉勇;陳年生;;遺傳算法和神經(jīng)網(wǎng)絡(luò)耦合的金融預(yù)測(cè)系統(tǒng)[J];上海交通大學(xué)學(xué)報(bào);2016年02期

3 李清;任朝陽(yáng);;上市公司會(huì)計(jì)舞弊風(fēng)險(xiǎn)指數(shù)構(gòu)建及預(yù)警研究[J];西安交通大學(xué)學(xué)報(bào)(社會(huì)科學(xué)版);2016年01期

4 盧馨;李慧敏;陳爍輝;;高管背景特征與財(cái)務(wù)舞弊行為的研究——基于中國(guó)上市公司的經(jīng)驗(yàn)數(shù)據(jù)[J];審計(jì)與經(jīng)濟(jì)研究;2015年06期

5 宋彪;朱建明;李煦;;基于大數(shù)據(jù)的企業(yè)財(cái)務(wù)預(yù)警研究[J];中央財(cái)經(jīng)大學(xué)學(xué)報(bào);2015年06期

6 廖小蘭;;管理層會(huì)計(jì)舞弊動(dòng)機(jī)與防范對(duì)策[J];財(cái)會(huì)通訊;2015年16期

7 陳佳聲;;上市公司、審計(jì)師與監(jiān)管機(jī)構(gòu)的財(cái)務(wù)舞弊博弈研究[J];審計(jì)研究;2014年04期

8 王澤霞;沈佳翔;甘道武;;上市公司管理舞弊風(fēng)險(xiǎn)因子探索——基于問(wèn)卷調(diào)查與因子分析[J];會(huì)計(jì)之友;2014年01期

9 沈淑霞;李潔萍;;上市公司管理舞弊審計(jì)的風(fēng)險(xiǎn)控制[J];中國(guó)管理信息化;2013年06期

10 郭毅夫;權(quán)思勇;;基于神經(jīng)網(wǎng)絡(luò)的創(chuàng)新型企業(yè)財(cái)務(wù)危機(jī)預(yù)警研究[J];統(tǒng)計(jì)與決策;2013年04期

相關(guān)碩士學(xué)位論文 前2條

1 周偉;EVA與神經(jīng)網(wǎng)絡(luò)相結(jié)合的財(cái)務(wù)預(yù)警模型研究[D];山東財(cái)經(jīng)大學(xué);2012年

2 王晶晶;我國(guó)上市公司財(cái)務(wù)報(bào)表舞弊識(shí)別研究[D];暨南大學(xué);2008年

，

本文編號(hào)：2345350