In Table 1, the correlations between different listing variables andthe ‘status’ dummy variable is presented. The pair wise correlationtest was performed with all reasonable variables. Some creditinformation variables were omitted if they suffered from a smalldata sample or if they were too much alike other variables. Thevariables that the borrower can have an influence on have beenpresented in the first four rows of the table. The rest of the rows arecredit information variables, which the borrower cannot influence atleast in the short-run.As a whole, the correlations are relatively small. There are a fewlogical reasons for this. First of all, we have used all the data available.This enables us to see the big picture, but for example the starting rateis very sensitive to the credit grade. For example, a 15% starting ratemight guarantee the success of the listing for an AA grade borrower,but the same starting rate might be too low for an HR grade borrowerto get her listing funded. Therefore, in the full data set the correlationsare lower than when examined one credit grade at a time. Againparticularly the starting rate, i.e. “the price of the loan”, is verysensitive to common market interest rates and risk premiums. Wehave used data from the full two and a half years of time. During thistime the federal interest rate has varied between 2 and 5.25% (FederalReserve [7]). In addition, the recent credit crisis has increased the riskpremiums substantially. Therefore, the correlations would be higher ifwe would look at data from shorter periods of time, where the marketfundamentals would be similar for all listings.The correlation analysis done with the full data set does enable usto compare the significance of different variables. As we can see, thecredit information variables have generally higher correlations thanthe decision variables. This is quite logical, as people with low creditgrades have difficulties in obtaining a loan no matter how high, forexample, the starting rate is. All the correlations are statisticallysignificant, because of the high number of observations. The ‘amountrequested’ and ‘starting rate’ have higher correlations than the‘funding option’ and the ‘duration’. The signs of these correlationsare in line with [9]. A higher starting rate increases the borrowerschances of getting the loan funded (note that Prosper.com auctionmechanism is reversed in the sense that high interest rate is bad forthe borrower, i.e. the seller, and good for bidders). Logically, a higheramount requested decreases the borrower's chances of having asuccessful listing. The funding option “Open for duration”, entered as1 increases the borrower's success probability, as is the case with thelonger duration.Next to the credit grade, the delinquency related variables havethe second highest correlation. The ‘current delinquencies’ seem to bethe most influential of these variables. The ‘homeownership’ showssome correlation and the correlation of ‘debt-to-income ratio’ isrelatively low. This is the case with the variable ‘income’ as well. Allthe signs of the variables are logical.In Table 2 the correlation analysis
在表1中,不同的列表變量和之間的相關性<br>的“狀態”虛擬變量被呈現。該成對相關<br>測試與所有合理的變量進行。一些信用<br>,如果他們從一個小遭遇信息變量被省略<br>的數據樣本或者如果他們太相像其他變量。該<br>變量的借款人可以對已經產生影響<br>的前四排桌子的呈現。該行的其餘部分是<br>信貸信息變量,借款人不能影響<br>在短期內至少。<br>作為一個整體,相關性相對較小。有幾個<br>這種邏輯的理由。首先,我們使用所有可用數據。<br>This enables us to see the big picture, but for example the starting rate<br>is very sensitive to the credit grade. For example, a 15% starting rate<br>might guarantee the success of the listing for an AA grade borrower,<br>but the same starting rate might be too low for an HR grade borrower<br>to get her listing funded. Therefore, in the full data set the correlations<br>are lower than when examined one credit grade at a time. Again<br>particularly the starting rate, i.e. “the price of the loan”, is very<br>sensitive to common market interest rates and risk premiums. We<br>have used data from the full two and a half years of time. During this<br>time the federal interest rate has varied between 2 and 5.25% (Federal<br>預訂[7])。此外,最近的信貸危機,增加了風險<br>大幅溢價。因此,相關性會更高,如果<br>我們看一下數據,從較短的時間內,在市場<br>基本面將是所有目錄類似。<br>與完整數據集進行相關性分析確實讓我們<br>比較不同變量的重要性。正如我們所看到的,<br>信用信息的變量比通常更高的相關<br>決策變量。這是很符合邏輯,因為人們與低信用<br>等級在獲得貸款,無論有多高,對於困難的<br>例子,起始速度。所有的結果都是統計學<br>顯著因為高若干意見。在'量<br>要求“和”開動率“具有比更高的相關<br>的融資選項”和“持續時間”。這些相關性的符號<br>是與[9]線。在更高的起點率增加了借款人<br>獲得資助的貸款(注意Prosper.com拍賣的可能性<br>機制在這個意義上扭轉了高利率是壞的<br>借款人,即賣家,和良好的投標人)。從邏輯上講,更高的<br>要求量減少有對借款人的機會<br>成功上市。這些資金選項“打開的時間”,輸入為<br>1增加借款人的成功概率,因為是用的情況下,<br>持續時間較長。<br>旁邊的信用等級,拖欠相關變量具有<br>第二高的相關性。“當前拖欠”似乎是<br>最有影響力的這些變量。在'住房擁有將會顯示<br>一定的相關性和'債務收入比'的相關性是<br>相對低的。這與可變收入“,以及情況。所有<br>變量的標誌是符合邏輯的。<br>表2中的相關分析
正在翻譯中..
