Regression models evaluation - metrics

 Yes! Apart from RMSE, you can use multiple other evaluation metrics depending on your problem type (regression or classification). Since you're working on a regression problem, here are some key alternatives:

---

📌 1. Mean Absolute Error (MAE)

$$MAE = \frac{1}{n} \sum_{i=1}^{n} |y_i - \hat{y_i}|$$

✅ Measures the average absolute difference between actual and predicted values.
✅ Less sensitive to large errors compared to RMSE.
🚫 Doesn’t penalize large errors as heavily as RMSE.

📌 Use When: You want a simple, interpretable metric without giving more weight to large errors.

---

📌 2. Mean Squared Error (MSE)

$$MSE = \frac{1}{n} \sum_{i=1}^{n} (y_i - \hat{y_i})^2$$

✅ Squaring the errors gives more weight to larger errors.
✅ Helps identify models that make large mistakes.
🚫 Hard to interpret since it’s in squared units.

📌 Use When: You want to penalize large errors more severely.

---

📌 3. Mean Absolute Percentage Error (MAPE)

$$MAPE = \frac{1}{n} \sum_{i=1}^{n} \left| \frac{y_i - \hat{y_i}}{y_i} \right| \times 100$$

✅ Expresses error as a percentage, making it easy to interpret.
🚫 Can be unreliable if actual values are very close to zero.

📌 Use When: You need a scale-independent metric (e.g., comparing performance across different datasets).

---

📌 4. R-Squared (R²) – Coefficient of Determination

$$R^2 = 1 - \frac{\sum (y_i - \hat{y_i})^2}{\sum (y_i - \bar{y})^2}$$

✅ Measures how well the model explains variance in the data.
✅ Values range from 0 to 1 (closer to 1 is better).
🚫 Doesn’t tell you how good your predictions are in absolute terms.

📌 Use When: You want to know how well your model fits the data overall.

---

📌 5. Adjusted R-Squared

✅ Similar to R², but adjusts for the number of predictors in the model.
🚫 Helps avoid overfitting in complex models with many variables.

📌 Use When: You have multiple independent variables and want a more reliable goodness-of-fit measure.

---

🔍 Choosing the Right Metric

Metric	Interpretation	Best Used When
RMSE	Penalizes large errors	General regression models
MAE	Absolute error, easier to interpret	When you need simplicity
MSE	Penalizes large errors more	Handling outliers
MAPE	Expresses error in %	When target values vary widely
R²	Measures model fit	When checking variance explanation
Adjusted R²	Improves R² for multiple features	When avoiding overfitting

---

🛠 Example: Calculating These Metrics in Python

python
from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score

# Example data
y_true = [100, 200, 300, 400, 500]
y_pred = [110, 195, 290, 405, 495]

# Compute metrics
mae = mean_absolute_error(y_true, y_pred)
mse = mean_squared_error(y_true, y_pred)
rmse = mean_squared_error(y_true, y_pred, squared=False)
r2 = r2_score(y_true, y_pred)

# Print results
print(f"MAE: {mae}")
print(f"MSE: {mse}")
print(f"RMSE: {rmse}")
print(f"R² Score: {r2}")

Would you like help choosing the best metric for your dataset? 🚀