Coefficient of Determination Math Example 4

Follow the full solution, then compare it with the other examples linked below.

Example 4

hard
A model has R2=0.95R^2=0.95. A researcher concludes 'the model is perfect and ready for deployment.' Identify two potential problems with this conclusion.

Solution

  1. 1
    Problem 1: R2=0.95R^2=0.95 is in-sample — may be due to overfitting; out-of-sample performance may be much lower
  2. 2
    Problem 2: High R2R^2 doesn't mean model assumptions are satisfied — residuals may be heteroscedastic, non-normally distributed, or show patterns indicating non-linearity
  3. 3
    Additional check: need residual plots, out-of-sample validation, and assumption diagnostics before deploying any model

Answer

Problems: (1) may be overfitting; (2) assumptions may be violated. R² alone insufficient for deployment decision.
High R2R^2 is necessary but not sufficient for a good model. It doesn't guarantee good out-of-sample predictions, satisfied assumptions, or causal validity. Model validation must include out-of-sample testing and residual diagnostics.

About Coefficient of Determination

The proportion of the total variation in the response variable yy that is explained by the linear relationship with the explanatory variable xx. It equals the square of the correlation coefficient: r2r^2.

Learn more about Coefficient of Determination →

More Coefficient of Determination Examples

Example 1 medium

A regression model has [formula] (total variation) and [formula] (unexplained variation). Calculate

Example 2 hard

Two models predict house prices: Model 1 (size only): [formula]. Model 2 (size + neighborhood + age)

Example 3 easy

The correlation between study hours and test score is [formula]. Calculate [formula] and interpret i

← All Coefficient of Determination Examples Coefficient of Determination Concept