Regression Analysis in English
Regression Analysis (Phân tích hồi quy) is a fundamental statistical technique used to understand and quantify how changes in independent variables affect a dependent variable.
Core Formula
Simple Linear Regression: Y = a + bX + ε
- Y — Dependent variable (outcome)
- X — Independent variable (predictor)
- a — Intercept (value of Y when X = 0)
- b — Coefficient (change in Y per unit change in X)
- ε — Error term
Types
| Type | Use Case |
|---|---|
| Simple Linear | One predictor variable |
| Multiple Linear | Two or more predictors |
| Logistic | Binary outcome (e.g., default/no-default) |
| Polynomial | Non-linear relationships |
Applications in Banking & Credit
- Credit risk scoring: Predicting probability of loan default.
- Interest rate modeling: Identifying macroeconomic drivers of lending rates.
- Revenue forecasting: Projecting SME growth trajectories.
Model quality is assessed using R² (coefficient of determination), p-values, and RMSE.
Practical Example
A lender analyzing SME loan defaults might run a logistic regression using variables such as revenue growth, debt-service coverage ratio, and years in operation to estimate the probability that a borrower defaults within 12 months. If the model shows a coefficient of -0.8 for revenue growth, this indicates that higher revenue growth is strongly associated with lower default risk, helping credit teams calibrate approval thresholds and pricing more objectively than relying on judgment alone.
Vay vốn kinh doanh
Hiểu rõ thuật ngữ tài chính giúp doanh nghiệp chuẩn bị hồ sơ vay vốn tốt hơn và tăng khả năng được phê duyệt khoản vay. Đăng ký so sánh lãi suất từ nhiều ngân hàng qua BizLoan.