5a: and other applications of scoring methodology
5a.1 and 1988 Capital Accord
5a:巴塞尔协议等评分方法的应用
Copyright By PowCoder代写 加微信 powcoder
5。巴塞尔委员会和1988年资本协定
主要(十国集团)工业国的中央银行/银行监管机构
每三个月在巴塞尔的国际清算银行见面一次
Central Banks/Bank regulators of major (G10) industrial countries
Meet every 3 months at Bank for International Settlements in I Capital Accord 19881988年《巴塞尔资本协议》
其目的是设定最低监管资本金要求，以确保银行能够在任何时候返还储户资金
资本比率=可用资本/风险加权资产
资本充足率为> 8%
资本是股东权益和留存收益(一级资本);其他资源(第2层)
Aim is to set up minimum regulatory capital requirements in order to ensure that banks are able, at all times, to give back depositor’s funds
Capital ratio=available capital/risk-weighted assets
Capital ratio should be > 8%
Capital is shareholder equity and retained earnings (tier 1); other resources (tier 2)
Need for new Accord需要新的协议
Regulatory arbitrage
Not sufficient recognition of collateral guarantees
Solvency of debtor not taken into account
Fixed risk weights per asset class (‘standardised approach’)
未考虑债务人的偿付能力
每类资产的固定风险权重(“标准化方法”)
5a.2 Three pillars of I
Treatment of Credit Risk
Standard Approach
Internal Ratings Based Approach
Foundation
Treatment of Operational Risk
Treatment of Market Risk
Calculation of IRB granularity adjustment to
IRB粒度调整到资本的计算
Sound internal processes to evaluate risk
Supervisory monitoring
健全的内部流程来评估风险
Semi-annual disclosure
Bank’s risk profile
Qualitative and Quantitative information
Risk management processes
Risk management Strategy
Objective is to inform potential investors
目的是通知潜在投资者
Pillar 1: Minimum Capital
Requirement最低资本要求
Pillar 2: Supervisory Review
Process监督检查过程
Pillar 3: Market Discipline and
Public Disclosure市场纪律及公开披露
IRB (Internal Ratings Based) approaches for credit risk in corporate and retail exposures
IRB(基于内部评级的)公司和零售风险的信用风险评估方法
Portfolio segmented; risk assessed at “segment” level
For each segment need
PD, probability of default
LGD, (% of loan lost given default)
DLGD is LGD in downturn conditions
EAD: exposure at default
For corporates: M maturity (Macaulay duration http://www.investopedia.com/terms/m/macaulayduration.asp ) of loan
Foundation approach –
Estimate own PD. use regulators LGD, EAD, M estimates
Advanced approach corporates
Estimate own PD, LGD, EAD and M
Retail -only advanced approach allowed
Estimate own PD, LGD and EAD
Expected losses in a segment is EL= PD.LGD. EAD
Must be covered by profits, so factored into pricing
is concerned with Unexpected Losses, UL in a segment
UL = K.EAD =(PD). (D)LGD. EAD
投资组合分段;按“部门”水平评估风险
针对每个细分市场的需求
LGD(贷款损失的百分比)
DLGD是低迷时期的LGD
对于公司:M贷款期限(Macaulay duration http://www.investopedia.com/terms/m/macaulayduration.asp)
估计自己的PD。使用监管机构LGD, EAD, M估计
估计自己的PD, LGD, EAD和M
估计自己的PD, LGD和EAD
一个部门的预期损失是EL= PD.LGD。含铅
The I Value at Risk (VAR) model (contd.)巴塞尔协议II风险价值(VAR)模型(contd.)
Credit Loss
Expected Loss
Unexpected Loss
Provisions
Basel sets α* at 99.9%, meaning that there is a 0.1% chance (once in 1000 years)
that an institution’s capital would fail to absorb the unexpected loss and becomes insolvent!
巴塞尔集α*在99.9%,这意味着有0.1%的可能性(在1000年),一个机构的资本无法吸收意外损失和破产!
Results of QIS5(Quantitative Impact Studies): Where the savings are made QIS5(定量影响研究)的结果:节省的地方
https://eiopa.europa.eu/Publications/QIS/QIS5-technical_specifications_20100706.pdf
This is to test the actual implementation of solvency standard under Basel. MRC refers to the minimum required capital. Lots of resources to provide details on EBA website: https://www.eba.europa.eu/.这是为了检验巴塞尔协议下偿付能力标准的实际执行情况。MRC是指所需的最低资本。许多资源提供详细的EBA网站:https://www.eba.europa.eu/。
Credit risk weighted assets for
corporate and retail exposures信用风险加权资产，用于公司和零售风险
Capital needed is
where N is Cumulative Normal Distribution, N-1 is inverse distribution and R is correlation
Only covers unexpected risk; so if R=0, K=0 ; if R=1, K=LGD(1-PD)
其中N为累积正态分布，N-1为负态分布，R为相关关系
只承保意外风险;如果R=0 K=0;如果R = 1, K =乐金显示器(1 pd)
Retail exposures
M=1 ( maturity term disappears)
For Mortgages R=0.15
For Revolving R=0.04
For other retail
M=1(到期期限消失)
抵押贷款R = 0.15
对于旋转R = 0.04
Corporate exposures
b=(.11852-.05478ln(PD))2
Idea behind Basel formula巴塞尔公式背后的理念
Uses Vasicek one factor extension of Merton corporate credit risk model (1974)
Merton model: default occurs when debts exceed assets
Vasicek extension: movement in assets is sum of idiosyncratic effect and that due to common factor ( world economy)
使用Vasicek对Merton公司信用风险模型的单因素扩展(1974)
默顿模型:当债务超过资产时发生违约
Vasicek扩展:资产的移动是特殊效应和共同因素(世界经济)的总和。
Resources:
http://www.bbk.ac.uk/ems/for_students/msc_finEng/pricing_emms014p/ab8.pdf;
http://www.northinfo.com/documents/568.pdf;
http://faculty.chicagobooth.edu/pietro.veronesi/research/CNV_EmpiricalMertonModel_PreliminaryVersion.pdf;
https://www.soa.org/library/newsletters/risk-management-newsletter/2009/june/jrm-2009-iss16-wang.pdf)
Idea is to fit model so if economy is at its long run average the default rate is PD of segment- then calculate what is PD if economy has 1 in 1000 year downturn ( .999 level)
Input long run PD –> output (PD) , PD in downturn
Correlation between defaults is given through correlation with world economy
If R=1 then if one fails they all fail
If R= 0 then defaults are independent
这是为了适应模型，如果经济处于长期平均水平，违约率是部门的PD，然后计算PD，如果经济在1000年的衰退中有1(。999水平)
长期 Input PD –> 输出 (PD) PD 在 衰退
违约之间的相关性是通过与世界经济的相关性得到的
如果R=1，那么如果一个失败，它们都失败
如果R= 0，则默认值是独立的
5a.3 Is I model appropriate for portfolios of consumer loans?巴塞尔协议II模型适用于消费者贷款组合吗?
corporate loans
well established market
market price continuously available
bonds only infrequently withdrawn
市场价格持续有效的债券只是很少被撤回
contingent claim model says default is when loans exceeds assets
或有索赔模型认为，违约是指贷款超过资产
Correlation between defaults related to correlation of assets related to correlation of share prices
Economic conditions built into models
相关的;相关的;相关的违约与资产相关与股价相关
consumer loans
no established market-only securitization sales- these badly priced消费者贷款没有既定的市场——只有证券化销售——这些定价糟糕的
no price available as no public sales无价格可供选择，因为没有公开销售
consumers often leave lender (attrition)
default caused by cash flow (consumer has no idea of assets nor can realise them)消费者往往会因为现金流(消费者对资产没有概念，也无法变现)导致贷款人(自然损耗)违约
no share price surrogate for correlation of defaults没有股价可以替代违约相关性
Economic conditions not in models
But one could reinterpret corporate loans to be more
acceptable for consumer lending ( de Andrade)
但是人们可以重新解释公司贷款，使之更能被消费者贷款所接受(温德林·穆尼亚兹·安德拉德)
Conclusions of
Proposal ensures credit scoring of retail portfolios is essential for any bank which wants to profitably use IRB approach on any loan portfolio
Higher profile with concomitant opportunities and threats
建议确保零售组合的信用评分对任何希望在任何贷款组合上使用IRB方法获利的银行都是至关重要的
更高的姿态，伴随的机会和威胁
Calibration of credit scoring needs to be strengthened
Not just rank consumers correctly in terms of default risk but scores must give accurate default probabilities
信用评分的校准需要加强
不仅要根据违约风险对消费者进行正确的排名，而且评分必须给出准确的违约概率
Identified that there are not good models of credit risk of portfolios of credit识别出没有好的信用风险模型的信用组合
Need to relate Point in Time PD toThrough the Cycle PD需要将时间点PD与周期PD相关联
Need to model Loss given Default especially for unsecured loans需要模型的损失，特别是为无担保贷款违约
Since the global financial crisis , BIS (Bank of International Settlements https://www.bis.org/) have issued a new Accord in late 2010 to start in 2013 and fully implemented by 2019.
Main changes are how the capital is held
I says 25% in core Tier 1 ( equity /retained earnings); 50% are Tier 1
II says 57% in core Tier 1: 75% in Tier 1
Capital Conservation Buffer
keep another 32% of MCR as core Tier 1 capital
to absorb losses in times of financial stress. As it gets used up less dividends allowed
Countercyclical capital buffer
Another up to 32% of MCR
to be built up in good times ready for stress
Restrictions on dividends, bonuses if not full amount kept
自全球金融危机以来，国际清算银行(Bank of International Settlements https://www.bis.org/)于2010年底发布了一项新协议，将于2013年开始实施，并于2019年全面实施。
主要的变化是如何持有资本
巴塞尔协议II规定25%的核心一级资本(股本/留存收益);50%属于第一层
巴塞尔协议III规定，核心一级资本的57%:一级资本的75%
资本保存缓冲保持另外32%的MCR作为核心一级资本
在金融压力下吸收损失。当它用完的时候，允许的股息就少了
另一个高达32%的MCR将建立i
Recognition no satisfactory model for credit risk of portfolios of consumer loans came from
requires estimates of credit risk of subportfolios of consumer loans
Forced to use corporate credit risk Merton-Vasicek model
Sub prime mortgage crisis of 2007/8
correlations between mortgage defaults not correctly modelled
Pricing of securitization products needed portfolio level consumer credit risk model
In 3 months at end of 2007 Credit rating agencies (CRA) downgraded more RMBS than total downgrades of all products in previous 21 years
消费者贷款组合信用风险的识别没有令人满意的模型
巴塞尔协议要求对消费贷款的次级投资组合的信用风险进行评估
被迫采用企业信用风险Merton-Vasicek模型
2007/8年的次贷危机
抵押贷款违约之间的相关性没有被正确地建模
证券化产品定价需要投资组合层面的消费者信贷风险模型
在2007年底的3个月内，信用评级机构(CRA)对RMBS的降级超过了过去21年所有产品的全部降级
5a.5 Portfolio Level Consumer Credit Risk
投资组合层面的消费者信贷风险
Corporate Credit Risk Models企业信用风险模型
Three types of models well established
Structural Models: used in
Loan defaults if debts exceed assets
Correlation in assetsCorrelation in share price
Basel formula comes from simplified structural model ( Merton-Vasicek)
assets depending on one common factor ( world economy)
Consumer defaults if debts exceed “value of reputation”
Behavioural score is surrogate for value of reputation
, Thomas, Expert Systems and its applications 2007
Reduced Form Default Mode Models
Estimate time to default based on interest rate forecasts and ratings grade
Correlation is through common interest rate movements
Reduced Form Mark to market Models
Estimate time to move to other rating grade, including default
Also estimate transition matrix of movements between grades
Transitions depend on economic climate
Consider consumer loan portfolio equivalent of Reduced form default model
Survival analysis approach- proportional hazards model
Use behavioural score bands ( risk grades) and economic variables
三种类型的模型建立良好
结构模型:在巴塞尔协议中使用
如果债务超过资产，贷款违约
在 assetsCorrelation Correlation 股价
巴塞尔公式来自简化结构模型(Merton-Vasicek)
取决于一个共同因素的资产(世界经济)
如果债务超过“信誉价值”，消费者违约
行为分数是声誉价值的替代物
Thomas ，《专家系统及其应用》2007
根据利率预测和评级等级估计违约时间
5a.6 What caused the Credit crunch?
Frictions caused by securitization leading to subprime mortgage crisis ( Ashcraft, Schuermann FedRes 2008)
是什么导致了信贷紧缩?证券化引起的摩擦导致次贷危机(Ashcraft, Schuermann FedRes 2008)
Frictions that caused the sub prime crisis
造成了次贷危机的摩擦
1.Predatory Lending: Originator v borrower
Subprime borrowers financially unsophisticated ;Originators not explain terms of loan
2. Mortgage Fraud: Originator v Arranger
Originator knows more about borrower than arranger
3. Adverse selection: arranger v third parties
Arranger can securitize bad loans and keep the good ones
4. Moral Hazard : Servicer v borrower
Borrower should look after house , insure it etc; in delinquency not incentive to do that
5. Moral hazard; servicer v third parties
Servicer makes money while loan is on books and can inflate expenses for its services ; investor has no check on this
6. Principal agent: asset manager v investor
Investor puts in his money but the asset fund manager makes the decision , maybe to maximise his pay/bonus
7. Model error; Credit rating agency v investor
Ratings agency paid by arranger not by investor
Fed Reserve felt sub prime caused by 1,2,3,6 and 7
1.掠夺性贷款:发起人对借款人
次级借款人在财务上不成熟;发起人不解释贷款条款
2. 抵押贷款欺诈:发起人v安排者
发起人比安排者更了解借款人
3.逆向选择:安排v第三方
安排者可以将不良贷款证券化并保留好贷款
4. 道德风险:服务方vs借方
借款人应看家、投保等;拖欠，没有动机去做
5. 道德风险;服务方对第三方
服务赚钱，而贷款在账面上，可以膨胀费用为其服务;
6. 委托代理人:资产管理人vs投资者
投资者投入资金，但资产基金经理做出决定，或许是为了让自己的薪酬/奖金最大化
7. 模型误差;信用评级机构v投资者
评级机构由安排人支付，而非投资者
SEC Review of Credit Rating Agencies ( July 08)
美国证券交易委员会信贷评级机构检讨(2008年7月)
CRA Modelling processCRA建模过程
Estimate估计
Probability of default违约概率
Loss severity损失严重程度
Cash flow现金流
Using loan/borrower characteristics at time of originations ( given by originator)
Different models for probability of default used by three agencies
One used hazard rate type model for default and allowed for prepayment
One used logistic regression
One used static factor model
Until 2008 no special model for sub prime loans
Monte Carlo simulation of performance of model under different economic conditions
Adjustments made by comparing subjective risk of portfolio compared with previously issued portfolios
Information used was ; Credit score at origination; loan term; interest rate; for rent or occupation
在发放贷款时使用贷款/借款人特征(由发放人提供)
三个机构使用不同的违约概率模型
一种是违约风险率类型模型，允许提前支付
一种使用静态因素模型
直到2008年才出现了针对次级贷款的特殊模式
蒙特卡罗仿真模型在不同经济条件下的性能
通过将投资组合的主观风险与之前发行的投资组合进行比较做出的调整
SEC Review of Credit Rating Agencies ( July 08)
美国证券交易委员会信贷评级机构检讨(2008年7月)
Identified a number of failings in CRAs approach to rating RMBS and CDO(collateralized debt obligations)
Conflict of interest利益冲突
Issuer pays发行人支付
12 originators account for 80% of RMBS12个发起人占RMBS的80%
Growth in RMBS rating outstripped growth in analysts
Historic Data历史数据
Parameters of model updated very infrequently
Increase in subprime from $96 billion in 96 to $600 billion in 2006
2/28 mortgages were 31% in 99 and 69% in 05
40 year mortgages 0% of subprime in 05 , 27% in 06
Monitoring of models
No loan level monitoring
Triggers were remaining over collateralization ( i.e. dollar losses)
CDOs considered as RMBS or corporate loans
CDOs used similar models as RMBS but RMBS have 50+ inputs , CDOs had 5- credit rating, maturity, asset type, country, industry
Assumed bad rates of CDOs given by bad rates of corporate bonds
No history of CDOs based on subprime RMBS
识别出CRAs对RMBS和CDO(债务抵押债券)评级方法的若干缺陷
RMBS评级的增长超过了分析师的增长
模型参数更新非常不频繁
次贷从96年的960亿美元增加到2006年的6000亿美元
99年2/28的抵押贷款占31%，05年占69%
40年期抵押贷款占05年次级抵押贷款的0%，06年占27%
触发点仍然是超额抵押(即美元损失)
被认为是RMBS或公司贷款的债务抵押债券
cdo使用与RMBS类似的模型，但RMBS有50+的投入，cdo有5-信用评级、期限、资产类型、国家、行业
假定债务抵押债券的坏利率是由公司债券的坏利率给出的
没有基于次级RMBS的债务抵押债券的历史
5a. 7 Non credit applications of scoring
Marketing谁会对报价做出反应。使用哪些营销渠道
who will respond to an offer. Which marketing channels to use
Tax inspection税务检查
Which tax forms to investigate; which firms to audit for VAT purposes调查哪些税务表格;哪些公司需要进行增值税审计
Payments of fines支付罚款
Child support Agency used it儿童支持机构使用了它
Parole .假释。谁会获得假释，谁会被关在监狱里
Who gets parole and who stays in prison
Intensive Care Assessment.重症监护的评估
Who is likely to recover from severe head trauma谁有可能从严重的头部创伤中康复
University recruitment大学招聘
Proteomics蛋白质组学
0.05 0.05 0.05 0.05
0.1 0.1 0.1 0.1
0.15 0.15 0.15 0.15
0.2 0.2 0.2 0.2
0.25 0.25 0.25 0.25
0.3 0.3 0.3 0.3
0.35 0.35 0.35 0.35
0.4 0.4 0.4 0.4
0.45 0.45 0.45 0.45
0.5 0.5 0.5 0.5
0.55 0.55 0.55 0.55
0.6 0.6 0.6 0.6
0.65 0.65 0.65 0.65
0.7 0.7 0.7 0.7
0.75 0.75 0.75 0.75
0.8 0.8 0.8 0.8
0.85 0.85 0.85 0.85
0.9 0.9 0.9 0.9
0.95 0.95 0.95 0.95
residential
other retail
Updated Basel capital requirements for K to cover UL ( LGD=0.5)
0.131752954
0.0486618776
0.0590357053
0.1332039191
0.1816982237
0.074571819
0.0671491611
0.1716328049
0.2095311545
0.0923422152
0.078756275
0.1969185425
0.224994511
0.1048755965
0.0891354323
0.211761419
0.2322484731
0.1135153805
0.0969284511
0.2192026373
0.2335553961
0.119035866
0.1022025698
0.2211702413
0.2303147617
0.1219376132
0.1052485976
0.2188858788
0.2234660223
0.1225676845
0.1063233331
0.2131735307
0.2136767489
0.1211777973
0.1056234031
0.2046201222
0.2014416717
0.1179557074
0.103296359
0.1936614416
0.1871394625
0.1130433815
0.0994524563
0.1806314185
0.1710674439
0.1065479045
0.0941726349
0.1657921625
0.1534638795
0.0985479215
0.0875131775
0.1493531408
0.1345228379
0.0890969001
0.0795076673
0.1314837974
0.11440436
0.0782235369
0.0701662627
0.1123219073
0.0932414497
0.065928629
0.0594713406
0.0919788145
0.0711446341
0.052175801
0.0473664438
0.0705418026
0.0482040146
0.0368676355
0.0337288415
0.0480723769
0.024486397
0.0197719063
0.0182851262
0.0245938522
0.0000000051
0.05 0.05 0.05
0.1 0.1 0.1
0.15 0.15 0.15
0.2 0.2 0.2
0.25 0.25 0.25
0.3 0.3 0.3
0.35 0.35 0.35
0.4 0.4 0.4
0.45 0.45 0.45
0.5 0.5 0.5
0.55 0.55 0.55
0.6 0.6 0.6
0.65 0.65 0.65
0.7 0.7 0.7
0.75 0.75 0.75
0.8 0.8 0.8
0.85 0.85 0.85
0.9 0.9 0.9
0.95 0.95 0.95
resid mortgages
Updated Basel capital requirements for K to cover UL ( LGD=0.5)
0.131752954
0.0486618776
0.0590357053
0.1816982237
0.074571819
0.0671491611
0.2095311545
0.0923422152
0.078756275
0.224994511
0.1048755965
0.0891354323
0.2322484731
0.1135153805
0.0969284511
0.2335553961
0.119035866
0.1022025698
0.2303147617
0.1219376132
0.1052485976
0.2234660223
0.1225676845
0.1063233331
0.2136767489
0.1211777973
0.1056234031
0.2014416717
0.1179557074
0.103296359
0.1871394625
0.1130433815
0.0994524563
0.1710674439
0.1065479045
0.0941726349
0.1534638795
0.0985479215
0.0875131775
0.1345228379
0.0890969001
0.0795076673
0.11440436
0.0782235369
0.0701662627
0.0932414497
0.065928629
0.0594713406
0.0711446341
0.052175801
0.0473664438
0.0482040146
0.0368676355
0.0337288415
0.024486397
0.0197719063
0.0182851262
RES MORG UL 0.15 EL+UL
LGD\PD 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1
0.05 0 0.0131752954 0.0181698224 0.0209531154 0.0224994511 0.0232248473 0.0233555396 0.0230314762 0.0223466022 0.0213676749 0.0201441672 0.0187139462 0.0171067444 0.0153463879 0.0134522838 0.011440436 0.009324145 0.0071144634 0.0048204015 0.0024486397 0 0 0.0631752954 0.1181698224 0.1709531154 0.2224994511 0.2732248473 0.3233555396 0.3730314762 0.4223466022 0.4713676749 0.5201441672 0.5687139462 0.6171067444 0.6653463879 0.7134522838 0.761440436 0.809324145 0.8571144634 0.9048204015 0.9524486397 1
0.1 0 0.0263505908 0.0363396447 0.0419062309 0.0449989022 0.0464496946 0.0467110792 0.0460629523 0.0446932045 0.0427353498 0.0402883343 0.0374278925 0.0342134888 0.0306927759 0.0269045676 0.022880872 0.0186482899 0.0142289268 0.0096408029 0.0048972794 0 0 0.0763505908 0.1363396447 0.1919062309 0.2449989022 0.2964496946 0.3467110792 0.3960629523 0.4446932045 0.4927353498 0.5402883343 0.5874278925 0.6342134888 0.6806927759 0.7269045676 0.772880872 0.8186482899 0.8642289268 0.9096408029 0.9548972794 1
0.15 0 0.0395258862 0.0545094671 0.0628593463 0.0674983533 0.0696745419 0.0700666188 0.0690944285 0.0670398067 0.0641030247 0.0604325015 0.0561418387 0.0513202332 0.0460391638 0.0403568514 0.034321308 0.0279724349 0.0213433902 0.0144612044 0.0073459191 0 0 0.0895258862 0.1545094671 0.2128593463 0.2674983533 0.3196745419 0.3700666188 0.4190944285 0.4670398067 0.5141030247 0.5604325015 0.6061418387 0.6513202332 0.6960391638 0.7403568514 0.784321308 0.8279724349 0.8713433902 0.9144612044 0.9573459191 1
0.2 0 0.0527011816 0.0726792895 0.0838124618 0.0899978044 0.0928993893 0.0934221584 0.0921259047 0.0893864089 0.0854706996 0.0805766687 0.074855785 0.0684269775 0.0613855518 0.0538091352 0.045761744 0.0372965799 0.0284578537 0.0192816058 0.0097945588 0 0 0.1027011816 0.1726792895 0.2338124618 0.2899978044 0.3428993893 0.3934221584 0.4421259047 0.4893864089 0.5354706996 0.5805766687 0.624855785 0.6684269775 0.7113855518 0.7538091352 0.795761744 0.8372965799 0.8784578537 0.9192816058 0.9597945588 1
0.25 0 0.065876477 0.0908491118 0.1047655772 0.1124972555 0.1161242366 0.1167776981 0.1151573808 0.1117330112 0.1068383745 0.1007208359 0.09
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com