Random forest regression results for predicting digital financial practices (DFP)
| Metric/Variable | Description | Result | Interpretation |
|---|---|---|---|
| Model type | Random Forest regression (ntree = 1,000) | – | Non-parametric ML approach for predictive validation |
| R2 (test set) | Coefficient of determination | 0.687 | Model explains ∼68.7% of the variance in DFP |
| RMSE (test set) | Root mean square error | 0.523 | Indicates good predictive accuracy |
| DFL (%IncMSE) | Increase in mean squared error when variable is permuted | 47.6% | DFL = most important predictor |
| Education level (%IncMSE) | 28.2% | 2nd strongest predictor | |
| Age (%IncMSE) | 26.2% | 3rd strongest predictor | |
| Income level (%IncMSE) | 19.3% | Moderate influence | |
| Gender (%IncMSE) | 7.5% | Least influence | |
| Partial dependence trend | DFL → Predicted DFP | Monotonic Positive | Confirms consistent behavioural improvement with higher literacy |
| Metric/Variable | Description | Result | Interpretation |
|---|---|---|---|
| Model type | Random Forest regression (ntree = 1,000) | – | Non-parametric ML approach for predictive validation |
| Coefficient of determination | 0.687 | Model explains ∼68.7% of the variance in DFP | |
| RMSE (test set) | Root mean square error | 0.523 | Indicates good predictive accuracy |
| DFL (%IncMSE) | Increase in mean squared error when variable is permuted | 47.6% | DFL = most important predictor |
| Education level (%IncMSE) | 28.2% | 2nd strongest predictor | |
| Age (%IncMSE) | 26.2% | 3rd strongest predictor | |
| Income level (%IncMSE) | 19.3% | Moderate influence | |
| Gender (%IncMSE) | 7.5% | Least influence | |
| Partial dependence trend | DFL → Predicted DFP | Confirms consistent behavioural improvement with higher literacy |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.