Loan Credit Risk

SQL Database Tables


Below are the different data sets that you will find in the Loans_R database after deployment.

Table Description
LoanRaw data about each loan
BorrowerRaw data about each borrower
MergedThe merged result of Loan and Borrower
StatsModes or Means of variables of Merged for missing values replacement
Merged_CleanedMerged table with missing values replaced
BinsSerialized list of cutoffs used to bin numeric variables
Merged_FeaturesAnalytical data set: cleaned merged table with new features
Column_InfoSerialized list of variable information including factors and their levels
Hash_IdLoan Ids and Mapping through hash function for splitting
ModelSerialized version of the trained logistic regression
Logistic_CoeffCoefficients of the logistic regression formula in decreasing order of magnitude
Predictions_LogisticPredictions made on the testing set
MetricsPerformance metrics of the evaluated model
Operational_MetricsPercentiles of the scores, corresponding score thresholds and bad rates among scores higher than thresholds
Scores_AverageAverage of the predicted scores used in scores transformation
ScoresScores computed on the testing set, transformed with Operational_Metrics