With enough variables, it is possible to get an $R ^ { 2 }$ value close to 1 , even if the variables have no particular meaning in the model

Question

Quizplus · Accepted Answer

Adding more variables to a regression model can artificially inflate the $R^2$ value because $R^2$ measures the proportion of the variance for a dependent variable that's explained by an independent variable or variables in a regression model. Even if the additional variables have no real predictive power or meaningful relationship with the outcome, their inclusion can still increase the $R^2$ value, leading to a model that appears to fit the data well even when it does not. This phenomenon is known as overfitting.

With Enough Variables, It Is Possible to Get An R2R ^ { 2 }R2

With Enough Variables, It Is Possible to Get An $R ^ { 2 }$