Dark Mode
Customer Loan Application Data
Finance & Banking Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Data on loan applications, featuring 1,000 records that detail applicants' financial status, credit history, and loan specifics. The primary objective for this dataset is to facilitate the analysis of credit risk patterns and the development of models to predict loan defaults.
Columns
checking_balance: The customer's current account balance in deutschmarks, categorised as '< 0 DM', '1 - 200 DM', '> 200 DM', or 'unknown'. 'Unknown' represents 39% of the entries.months_loan_duration: The duration of the loan in months. Values range from 4 to 72 months, with a mean duration of 20.9 months.credit_history: The applicant's credit history. 'Good' is the most frequent category at 53%, followed by 'critical' at 29%.purpose: The stated purpose of the loan. 'Furniture/appliances' is the most common purpose, accounting for 47% of loans, while 'car' accounts for 34%.amount: The loan amount, ranging from 250 to 18,400, with an average of 3,270.savings_balance: The savings account balance. '< 100 DM' is the most frequent category at 60%, with 'unknown' at 18%.employment_duration: The length of the applicant's employment. '1 - 4 years' is the most common duration at 34%, and '> 7 years' represents 25%.percent_of_income: The percentage of income allocated to loan repayment, with values between 1% and 4%.years_at_residence: The number of years at the current residence, with values ranging from 1 to 4 years.age: The applicant's age, spanning from 19 to 75 years, with an average age of 35.5 years.other_credit: Indicates the presence of other credit agreements. 'None' is the dominant category at 81%.housing: The applicant's housing status, such as 'own' (71%) or 'rent' (18%).existing_loans_count: The number of existing loans the applicant has, ranging from 1 to 4, with a mean of 1.41.job: The applicant's job type or classification. 'Skilled' is the most common type at 63%, and 'unskilled' at 20%.dependents: The number of dependents, which can be either 1 or 2.phone: Indicates the availability of a telephone. 40% of applicants have a telephone, while 60% do not.default: The target variable, signifying whether a loan defaulted ('yes' or 'no'). 30% of the loans in the dataset resulted in a default.
Distribution
The dataset comprises 1,000 records, provided in a CSV format. It features 17 columns, all of which are complete with 1000 valid entries and no missing values, ensuring data integrity.
Usage
- Developing predictive models for identifying loan default risks.
- Investigating the relationships between various financial indicators and overall credit risk.
- Deepening understanding of credit risk analysis methodologies.
Coverage
The data focuses on customer loan applications, utilising deutschmarks for financial figures, which suggests a context relevant to Germany. It includes demographic details such as applicant age (19-75 years), employment duration, and number of dependents. No specific time range for the data collection is provided.
License
CC BY-NC-SA 4.0 license
Who Can Use It
- Students and academic researchers: Ideal for academic projects, dissertations, and research studies focused on machine learning in finance or credit risk evaluation.
- Data scientists and analysts: Suitable for creating and testing models designed to forecast loan defaults and assess creditworthiness.
- Educators: An appropriate resource for teaching modules related to machine learning, financial analytics, or data science.
Dataset Name Suggestions
- Credit Risk Assessment Data
- Loan Default Prediction Records
- Banking Creditworthiness Dataset
- Customer Loan Application Data
Attributes
Original Data Source: Customer Loan Application Data
QUALITY