【摘要】：隨著經(jīng)濟(jì)一體化的高速發(fā)展,擔(dān)保作為一種很重要的籌集資金方式,在資金融通中被廣泛使用,但是由于其復(fù)雜且不透明特性,已成為加劇信貸風(fēng)險(xiǎn)暴露的重要原因,引起社會(huì)各界普遍關(guān)注。由于傳統(tǒng)分析方法對(duì)處理大規(guī)模擔(dān)保網(wǎng)絡(luò)問(wèn)題的局限性,復(fù)雜網(wǎng)絡(luò)等大數(shù)據(jù)技術(shù)則為擔(dān)保圈風(fēng)險(xiǎn)分析提供了契機(jī)。本文利用某金融機(jī)構(gòu)提供的擔(dān)保數(shù)據(jù),基于復(fù)雜網(wǎng)絡(luò)理論分析擔(dān)保網(wǎng)絡(luò)的拓?fù)浣Y(jié)構(gòu),重點(diǎn)識(shí)別擔(dān)保圈的社團(tuán)結(jié)構(gòu),以及揭示擔(dān)保風(fēng)險(xiǎn)的傳播機(jī)制,為有針對(duì)性的進(jìn)行風(fēng)險(xiǎn)防范,降低風(fēng)險(xiǎn)管理成本提供幫助。具體的研究?jī)?nèi)容可以歸納為以下幾方面:(1)提出一種快速的社團(tuán)挖掘算法Fast-SCAN(FSCAN)。SCAN算法的時(shí)間復(fù)雜度是O(4|E|2/|V|),針對(duì)該算法在大規(guī)模網(wǎng)絡(luò)社團(tuán)劃分中時(shí)間開銷較高的問(wèn)題,本文改進(jìn)后的時(shí)間復(fù)雜度是O(2|E|2/|V|-2K|E|/|V|),能夠高效發(fā)現(xiàn)大規(guī)模網(wǎng)絡(luò)中的社區(qū)結(jié)構(gòu)、集群點(diǎn)以及噪音點(diǎn)。(2)本文引入并查集思想,提出基于路徑壓縮的譜系劃分算法,可以發(fā)現(xiàn)擔(dān)保網(wǎng)絡(luò)中連通的擔(dān)保圈。結(jié)合圖論和復(fù)雜網(wǎng)絡(luò)理論對(duì)其結(jié)構(gòu)進(jìn)行分析,同時(shí)使用FSCAN算法識(shí)別核心擔(dān)保圈的社團(tuán)結(jié)構(gòu)。(3)針對(duì)待查信用主體在擔(dān)保網(wǎng)絡(luò)中隱藏的間接擔(dān)保關(guān)系,提出一種模擬鄰接表的數(shù)據(jù)結(jié)構(gòu),同時(shí)融合圖論的遍歷思想,提出待查信用主體K代關(guān)聯(lián)關(guān)系的路徑發(fā)現(xiàn)算法,發(fā)現(xiàn)待查信用主體之間隱藏的間接擔(dān)保關(guān)系,以防范擔(dān)保圈風(fēng)險(xiǎn)的爆發(fā)。(4)通過(guò)對(duì)擔(dān)保圈的風(fēng)險(xiǎn)傳播特性進(jìn)行分析,改進(jìn)災(zāi)害蔓延動(dòng)力學(xué)模型,提出更適用的風(fēng)險(xiǎn)傳播模型。仿真?zhèn)鞑ピ磳?duì)風(fēng)險(xiǎn)傳播的影響,對(duì)待查信用主體進(jìn)行信用評(píng)級(jí),并仿真關(guān)鍵企業(yè)在核心擔(dān)保群中風(fēng)險(xiǎn)傳播的路徑。最后,提出相應(yīng)監(jiān)管策略,為防范因擔(dān)保圈引發(fā)的金融危機(jī)提供理論參考。
[Abstract]:With the rapid development of economic integration, guarantee, as a very important way to raise funds, is widely used in financing, but because of its complex and opaque characteristics, it has become an important reason for increasing exposure to credit risk. Cause the general concern of all walks of life of the society. Because of the limitation of traditional analysis method to deal with the problem of large-scale guarantee network, big data technology such as complex network provides an opportunity for risk analysis of guarantee circle. Based on the theory of complex network, this paper analyzes the topological structure of the guarantee network by using the guarantee data provided by a financial institution, focuses on identifying the community structure of the guarantee circle, and reveals the spreading mechanism of the guarantee risk in order to guard against the risk. Help reduce risk management costs. The specific research contents can be summarized as follows: (1) the time complexity of a fast community mining algorithm Fast-SCAN (FSCAN). SCAN is O (4E _ 2 / V). In order to solve the problem of high time cost in mass network community partitioning, the time complexity of the improved algorithm is O (2E 2 / C-V 2K E / V), which can efficiently discover the community structure in large scale network. Cluster points and noise points. (2) in this paper, we introduce and search the idea of set, and propose a pedigree partition algorithm based on path compression, which can find the connected guarantee circle in the guaranteed network. Combined with graph theory and complex network theory, this paper analyzes its structure, and uses FSCAN algorithm to identify the community structure of the core guarantee circle. (3) aiming at the indirect guarantee relationship hidden by the credit subject in the guarantee network, In this paper, we propose a data structure that simulates the adjacency table, and combine the traversal idea of graph theory, and propose a path discovery algorithm for K-generation association relation of the credit subject to be checked, and find the hidden indirect guarantee relationship between the credit subject under check. (4) by analyzing the risk propagation characteristics of the guarantee circle, the dynamic model of disaster spread is improved, and a more suitable risk propagation model is put forward. This paper simulates the influence of the source of communication on the risk propagation, evaluates the credit rating of the credit subject, and simulates the path of risk transmission in the core guarantee group of key enterprises. Finally, the corresponding regulatory strategies are put forward to provide a theoretical reference for preventing the financial crisis caused by the guarantee circle.
【學(xué)位授予單位】：大連海事大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：O157.5

【相似文獻(xiàn)】

相關(guān)會(huì)議論文 前1條

1 郭小平;;“風(fēng)險(xiǎn)傳播”研究的范式轉(zhuǎn)換[A];2006中國(guó)傳播學(xué)論壇論文集（Ⅰ）[C];2006年

相關(guān)重要報(bào)紙文章 前1條

1 劉濤;公民行動(dòng)與風(fēng)險(xiǎn)傳播[N];中國(guó)社會(huì)科學(xué)院報(bào);2009年

相關(guān)博士學(xué)位論文 前1條

1 朱閔敏;肺結(jié)核風(fēng)險(xiǎn)傳播干預(yù)與效果評(píng)估[D];華中科技大學(xué);2013年

相關(guān)碩士學(xué)位論文 前10條

1 程宇婕;風(fēng)險(xiǎn)傳播理論研究[D];山西大學(xué);2010年

2 楊佳彤;媒體在科技風(fēng)險(xiǎn)傳播中的效應(yīng)研究[D];西北農(nóng)林科技大學(xué);2015年

3 王佳樂;茂名PX環(huán)境沖突事件：環(huán)境風(fēng)險(xiǎn)傳播中政府、公眾與媒體的角色互動(dòng)[D];陜西師范大學(xué);2015年

4 韓冰潔;基于WBS的復(fù)雜產(chǎn)品制造過(guò)程風(fēng)險(xiǎn)傳播與管理研究[D];江蘇科技大學(xué);2015年

5 裴思e，

本文編號(hào)：2309186