The那個 methodolgy方法論 or the或 sql for obtaining the data數據庫獲取數據 these curves from the private data export have already been explained in detail, but briefly…這些曲線的私人數據的出口已經詳細解釋，但簡單...
- The Y axis is the percentage of all loans orginated of a given age that are currently 1 month late or worse…在Y軸的百分比orginated所有貸款某一特定年齡，目前1月下旬或更糟...
- The X axis is the days loan origination.在X軸是天的貸款來源。
- The curves stop when there is less than 250 loans in the “bucket.”停止曲線時，小於250貸款的“水桶” 。
- These curves are for all loans of a given Credit Grade. One would get different curves (flatter slopes=better); for exmaple, if you were to choose loans with 2 or less inquires and 5 years of credit history you would see flatter slopes.這些曲線是所有貸款某一信用等級。之一將得到不同的曲線（平坦的斜坡=更好） ;舉例來說，如果您有選擇貸款2或更少詢問和5年的信用記錄就會看到平坦的山坡上。
The range is 7-8% for AA all the way to ~55% for HR.範圍是7-8 ％為機管局的所有道路〜 55 ％的人力資源。
(Click Graph for Larger Version) （點擊圖放大版）
Prosper Vintage Curves普羅斯珀經典曲線
Brief Explanation: These curves show the entire set of prosper loan broken down by credit grade and lined up along the x axis on their origination date… As a loan goes late (1 month or worse) it is counted (either by amount or by count) as late against the population… The curves stop when the loan population falls below 250 (ie there are 249 or less loans that age or older)…簡短的解釋：這些曲線顯示一整套繁榮細分貸款信用等級和排隊沿x軸的起始日期...作為貸款不用後期（ 1個月或更糟糕）這是個（無論數量或計數）晚對人口...曲線停止貸款時，人口低於250個（即有249個或更少的貸款歲或以上） ...
Recently a study最近的一項研究 from the University of Maryland claimed a peak default date around month 10 of a Prosper loan. This would translate into the largest delta in this graph over a month period. Does this graph confirm or deny that statement? Is it conclusive? Please leave a comment.來自馬里蘭大學的要求高峰預設的日期10個月左右的普羅斯珀貸款。這將轉化為最大的三角洲在這圖表超過三個月。這個圖證實或否認這種說法？是決定性的？請發表評論。
Here is the vintage curves by count (click graph for larger version)…這是老式曲線計數（點擊大圖版） ...
Here is the vintage curves by amount (larger loan go late at a higher rate and therefore on a percentage basis you would expect an increase), (click graph for larger version)…這是老式曲線的金額（較大的貸款去年底在一個較高的利率，因此，對一定比例的基礎你希望增加） ， （點擊大圖版） ...
Here is the SQL that I used to pull the underlying data out of the以下是SQL ，我來拉基本數據出 public and private data downloads公共和私人數據下載 … ...
DECLARE @DTD int宣告@ DTD的整數
SET @DTD=30集@ DTD的= 30
SELECT選擇
cast(aday-originationdate as int) as 'PIT',演員（ aday - originationdate為整數）作為'坑' ，
l.creditgrade, l.creditgrade ，
sum(PrincipalBalance+NetDefaults) as 'Amount',總結（ PrincipalBalance + NetDefaults ）作為'金額' ，
count(l.[key]) as 'Count',計數（湖[關鍵] ）作為'計數' ，
sum(case WHEN (mld.DPD!=0 and總和（時（ mld.DPD ！ = 0和
(mld.DPD+(aday-observationdate))>@DTD) THEN （ mld.DPD + （ aday - observationdate ） ） “ @ DTD的） ，然後
PrincipalBalance+NetDefaults ELSE 0 END) as 'AmountLate', PrincipalBalance + NetDefaults語句0完）作為' AmountLate ' ，
sum(case WHEN (mld.DPD!=0 and總和（時（ mld.DPD ！ = 0和
(mld.DPD+(aday-observationdate))>@DTD) THEN （ mld.DPD + （ aday - observationdate ） ） “ @ DTD的） ，然後
PrincipalBalance+NetDefaults ELSE 0 END)/ PrincipalBalance + NetDefaults語句0完） /
sum(PrincipalBalance+NetDefaults) as AmountLatePercentage,總結（ PrincipalBalance + NetDefaults ）作為AmountLatePercentage ，
sum(case WHEN (mld.DPD!=0 and總和（時（ mld.DPD ！ = 0和
(mld.DPD+(aday-observationdate))>@DTD) THEN （ mld.DPD + （ aday - observationdate ） ） “ @ DTD的） ，然後
1 ELSE 0 END) as 'CountLate',一日語句0完）作為' CountLate ' ，
sum(case WHEN (mld.DPD!=0 and總和（時（ mld.DPD ！ = 0和
(mld.DPD+(aday-observationdate))>@DTD) THEN （ mld.DPD + （ aday - observationdate ） ） “ @ DTD的） ，然後
1.0 ELSE 0.0 END)/count(l.[key]) as 'CountLatePercentage' 1.0語句0.0完） /伯爵（湖[關鍵] ）作為' CountLatePercentage '
FROM從
loan l貸款升
inner join creditprofile cp on cp.listingkey=l.listingkey內加入creditprofile蛋白的cp.listingkey = l.listingkey
inner join LoanPerformance mld on l.[key]=mld.loankey cross join alldays內加入LoanPerformance mld的湖[關鍵] = mld.loankey兩岸加入alldays
where哪裡
mld.observationdate = ( select top 1 observationdate mld.observationdate = （選取前1 observationdate
from LoanPerformance sub從LoanPerformance分
where sub.observationdate < aday在sub.observationdate “ aday
and sub.loankey=mld.loankey order by sub.observationdate DESC )和sub.loankey = mld.loankey秩序sub.observationdate降序）
and aday < getDate()和aday “ getDate （ ）
and aday >= '02/01/2006'和aday “ = '02 / 2006分之01 '
and l.creditgrade!='NC'和l.creditgrade ！ = '數控'
group by組
cast(aday-originationdate as int),演員（ aday - originationdate為整數） ，
l.creditgrade
having有
count(l.[key])>250 and計數（湖[關鍵] ） “ 250
sum(PrincipalBalance+NetDefaults)>0總結（ PrincipalBalance + NetDefaults ） “ 0
order by命令
'PIT' '坑' Related Articles 相關文章 Related Stores 相關店鋪 
