發(fā)布時(shí)間：2018-07-05 20:32

  本文選題：P2P網(wǎng)絡(luò)借貸 + 借款滿標(biāo)　； 參考：《河南大學(xué)》2017年碩士論文

【摘要】：隨著互聯(lián)網(wǎng)安全、第三方支付技術(shù)的日漸成熟、人們?nèi)粘Ｉ钪袑?duì)網(wǎng)上交易信任的增加,互聯(lián)網(wǎng)金融行業(yè)迅速發(fā)展。P2P網(wǎng)絡(luò)借貸作為典型的互聯(lián)網(wǎng)金融模式,為急需資金的個(gè)人和中小企業(yè)提供了融資渠道,其依靠第三方網(wǎng)絡(luò)平臺(tái),為個(gè)人與個(gè)人之間提供了公開(kāi)透明的小額信用交易的可能,也為實(shí)現(xiàn)金融脫媒和踐行普惠金融做出了有益貢獻(xiàn)。網(wǎng)絡(luò)借貸自2005年興起以來(lái),在全球范圍內(nèi)發(fā)展迅速,但也面臨著借貸成功率較低、資金成本過(guò)高等問(wèn)題。借款滿標(biāo)是指借款人在網(wǎng)貸平臺(tái)規(guī)定的期限內(nèi)其列表可得到足夠的出借人關(guān)注和投標(biāo),進(jìn)而足額籌集到所需資金的狀態(tài)。本文通過(guò)對(duì)國(guó)內(nèi)外文獻(xiàn)的梳理和分析,試圖從借款人信息出發(fā),探討投資人通過(guò)第三方網(wǎng)貸平臺(tái)進(jìn)行投資決策的過(guò)程,為借款人提高借款滿標(biāo)概率及投資者提高投資精度提供有益建議。借貸市場(chǎng)中的信息傳播方式及公開(kāi)程度對(duì)金融機(jī)構(gòu)的發(fā)展至關(guān)重要,在P2P網(wǎng)貸市場(chǎng)中交易雙方互不相識(shí)僅通過(guò)網(wǎng)絡(luò)產(chǎn)生聯(lián)系,因此借款人在關(guān)鍵信息方面通常更有優(yōu)勢(shì),大量潛在真實(shí)信息常常由借款人掌握,投資人則處于信息劣勢(shì),借貸雙方存在著明顯的信息不對(duì)稱。本文將信息不對(duì)稱理論應(yīng)用到網(wǎng)貸行業(yè),從網(wǎng)絡(luò)貸款平臺(tái)運(yùn)營(yíng)現(xiàn)狀入手,審視網(wǎng)貸經(jīng)營(yíng)模式,重現(xiàn)P2P網(wǎng)貸交易流程,以闡明借款人信息與借款滿標(biāo)之間的關(guān)系,并構(gòu)建借款滿標(biāo)概率預(yù)測(cè)模型。文中對(duì)國(guó)內(nèi)借貸網(wǎng)站——拍拍貸的借貸機(jī)制進(jìn)行了著重介紹,作為首家純信用無(wú)擔(dān)保網(wǎng)貸平臺(tái),拍拍貸從性質(zhì)上講最接近P2P網(wǎng)絡(luò)借貸的本質(zhì),因此選擇該平臺(tái)作為研究主體,同時(shí)也從該平臺(tái)運(yùn)用網(wǎng)絡(luò)爬蟲軟件選取實(shí)證分析的數(shù)據(jù)來(lái)源。文章首先對(duì)選取的數(shù)據(jù)進(jìn)行聚類分析,為了解不同特征的借款人使用平臺(tái)的效率,文章選取5個(gè)維度的指標(biāo),對(duì)全部樣本數(shù)據(jù)進(jìn)行聚類分析,分析結(jié)果將8個(gè)信用等級(jí)的借款人分為四類。從而得出評(píng)估授信狀況相似的借款人行為模式是有差異的,而信用等級(jí)不同的借款人之間也會(huì)存在共性,也即信用等級(jí)部分反映了借款人的違約風(fēng)險(xiǎn)進(jìn)而影響其借貸成功可能性。因此對(duì)于借款人的動(dòng)態(tài)行為規(guī)律,只以信用等級(jí)來(lái)界定是有失偏頗的。因此挖掘出信用等級(jí)這一信號(hào)所不能顯示的信息,識(shí)別用戶在平臺(tái)的表現(xiàn),可以找出借款人中出現(xiàn)的不同的行為模式,同時(shí)為借款滿標(biāo)概率的研究提供了契機(jī)。希望經(jīng)過(guò)借款滿標(biāo)概率預(yù)測(cè)模型的構(gòu)建,可以借助借款人在網(wǎng)貸平臺(tái)上披露的信息預(yù)測(cè)其借款滿標(biāo)可能性。接著,文章選取指標(biāo)以構(gòu)建借款滿標(biāo)概率預(yù)測(cè)模型,經(jīng)過(guò)對(duì)樣本數(shù)據(jù)進(jìn)行預(yù)處理及多重共線性診斷,得出自變量之間存在多重共線性。本文為解決解釋變量間存在的多重共線性問(wèn)題,以使最終構(gòu)建模型更加精確,在閱讀相關(guān)文獻(xiàn)后選擇使用主成分改進(jìn)的邏輯回歸方法。這種方法的核心在于經(jīng)過(guò)變換可將原本具有相關(guān)性的解釋變量綜合為少數(shù)幾個(gè)綜合指標(biāo),提取出的少數(shù)綜合指標(biāo)能反映原來(lái)多個(gè)變量的信息。最終以借款人信息為自變量,以借貸滿標(biāo)與否為因變量,依據(jù)抓取的數(shù)據(jù),經(jīng)過(guò)數(shù)據(jù)處理,用提取出的主成分代替原有全部解釋變量進(jìn)行邏輯回歸,用二元邏輯回歸方法構(gòu)建借款滿標(biāo)概率預(yù)測(cè)模型,之后再對(duì)各提取出的主成分進(jìn)行還原,來(lái)研究其他若干變量對(duì)滿標(biāo)概率的作用方向和影響程度。為借款人提高借款滿標(biāo)概率、投資人優(yōu)化投資決策提供參考。最后針對(duì)當(dāng)前借款成功率較低,信用機(jī)制不健全等現(xiàn)狀,提出相應(yīng)優(yōu)化對(duì)策及對(duì)未來(lái)的展望,并主要從完善社會(huì)征信體系、加強(qiáng)互聯(lián)網(wǎng)金融領(lǐng)域立法、健全網(wǎng)貸行業(yè)風(fēng)險(xiǎn)應(yīng)對(duì)機(jī)制三個(gè)方面給出建議。
[Abstract]:With the security of the Internet and the growing maturity of the third party payment technology and the increasing trust in online transactions in people's daily life, the Internet finance industry has rapidly developed.P2P network lending as a typical internet financial model, providing financing channels for individuals and small and medium enterprises, which are in urgent need of funds, and rely on the third party network platform as individuals. The possibility of open and transparent small credit transactions with individuals has also contributed to the realization of financial disintermediation and the implementation of Inclusive Finance. Since the rise of 2005, Internet lending has developed rapidly around the world, but it also faces the low success rate of borrowing and the higher funding. The loan full scale refers to the borrower in the net. In the time limit set by the loan platform, the list of the borrowers can get enough attention and bid, and then raise the state of the required funds in full. By combing and analyzing the literature at home and abroad, this paper tries to discuss the process of investment decision by the investor through the third party net loan platform to improve the borrower's full loan. The way and openness of information dissemination in the loan market is very important to the development of financial institutions. In the P2P net loan market, the non acquaintances of the two parties are only connected through the network, so the borrower is usually more advantageous in the key information and a large number of potential real information. In this paper, the information asymmetry theory is applied to the net loan industry. This paper applies the information asymmetry theory to the net loan industry, starts with the operation status of the network loan platform, examines the operation mode of the net loan, reproduces the P2P net loan transaction flow, in order to clarify the relationship between the borrower's information and the full standard of the loan. In this paper, the loan full standard prediction model is built and the loan mechanism of the domestic loan website - pat loan is introduced in this paper. As the first platform of the first pure credit unsecured net loan, the patting loan is most close to the nature of the P2P network loan from the nature, so the platform is chosen as the research subject and the network is also used on the platform. In order to understand the efficiency of the borrowers with different features, the article selects 5 dimensions to analyze the efficiency of the different features of the borrowers, and analyzes the data of all the samples and divides the 8 credit rating borrowers into four categories. The behavior patterns of borrowers with similar status are different, and the borrowers with different credit grades also have commonality, that is, the credit grade reflects the borrower's default risk and then affects the possibility of borrowing success. Therefore, the dynamic behavior law of the borrower is biased by the credit rating. It can identify the information that the credit rating can not show, identify the user's performance on the platform, find out the different behavior patterns in the borrower, and provide an opportunity for the study of the probability of the loan full scale. The information forecast its loan full scale possibility. Then, the article selects the index to construct the probability prediction model of the loan full scale, through the preprocessing of the sample data and the multiple collinear diagnosis, the multiple collinearity exists between the independent variables. This paper is to solve the multiply collinearity problem between the explanatory variables, so as to make the final construction model more The core of this method is to integrate the original explanatory variables with the original correlation into a few comprehensive indexes, and the extracted minority index can reflect the original variable quantity information. Finally, the borrower information is the independent variable, Taking the full standard of borrowing and lending as the dependent variable, according to the captured data, after data processing, the extracted principal component is used instead of all the original explanatory variables to carry out logic regression. The two element logic regression method is used to construct the loan full scale probability prediction model, then the main components of each extraction are reduced to study the full scale of the other variables. The direction and extent of the effect of the probability. It provides a reference for the borrower to improve the loan full standard probability and the investor to optimize the investment decision. Finally, it puts forward the corresponding optimization countermeasures and prospects for the future, aiming at the low success rate of the current borrowing and the imperfect credit mechanism and so on, and mainly from the good social credit system to strengthen the Internet financial field. Three suggestions are given to improve the risk response mechanism of the net loan industry.
【學(xué)位授予單位】：河南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：F724.6;F832.4

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 廖理;張偉強(qiáng);;P2P網(wǎng)絡(luò)借貸實(shí)證研究:一個(gè)文獻(xiàn)綜述[J];清華大學(xué)學(xué)報(bào)(哲學(xué)社會(huì)科學(xué)版);2017年02期

2 黃蓉;;基于聚類分析的數(shù)據(jù)挖掘方法研究[J];山東農(nóng)業(yè)大學(xué)學(xué)報(bào)(自然科學(xué)版);2017年01期

3 劉鏡秀;門明;;P2P網(wǎng)絡(luò)借貸市場(chǎng)對(duì)資本市場(chǎng)的風(fēng)險(xiǎn)溢出效應(yīng)[J];技術(shù)經(jīng)濟(jì);2016年11期

4 趙旭;周菁;趙子健;;中國(guó)P2P平臺(tái)借款成功率的影響因素研究[J];現(xiàn)代管理科學(xué);2016年05期

5 廖理;李夢(mèng)然;王正位;;聰明的投資者:非完全市場(chǎng)化利率與風(fēng)險(xiǎn)識(shí)別——來(lái)自P2P網(wǎng)絡(luò)借貸的證據(jù)[J];經(jīng)濟(jì)研究;2014年07期

6 朱晉川;;互聯(lián)網(wǎng)金融的產(chǎn)生背景、現(xiàn)狀分析與趨勢(shì)研究[J];農(nóng)村金融研究;2013年10期

7 宋文;韓麗川;;P2P網(wǎng)絡(luò)借貸中投資者出借意愿影響因素分析[J];西南民族大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年05期

8 王朋月;李鈞;;美國(guó)P2P借貸平臺(tái)發(fā)展:歷史、現(xiàn)狀與展望[J];金融監(jiān)管研究;2013年07期

9 陳冬宇;賴福軍;聶富強(qiáng);;社會(huì)資本、交易信任和信息不對(duì)稱——個(gè)人對(duì)個(gè)人在線借貸市場(chǎng)的實(shí)驗(yàn)研究[J];北京航空航天大學(xué)學(xué)報(bào)(社會(huì)科學(xué)版);2013年04期

10 馮軍政;陳英英;;P2P信貸平臺(tái):新型金融模式對(duì)商業(yè)銀行的啟示[J];新金融;2013年05期

相關(guān)碩士學(xué)位論文 前2條

1 張波;logistic回歸模型中自變量相對(duì)重要性的評(píng)價(jià)方法[D];寧波大學(xué);2012年

2 吳麗麗;基于Logistic回歸模型的商業(yè)銀行信用風(fēng)險(xiǎn)管理研究[D];哈爾濱工業(yè)大學(xué);2007年

，

本文編號(hào)：2101691