diff --git a/doc/modules/linear_model.rst b/doc/modules/linear_model.rst
index f0c0f94d61980..3b8e956136bf8 100644
--- a/doc/modules/linear_model.rst
+++ b/doc/modules/linear_model.rst
@@ -369,7 +369,7 @@ Mathematically, it consists of a linear model trained with a mixed
 :math:`\ell_1` :math:`\ell_2`-norm for regularization.
 The objective function to minimize is:
 
-.. math::  \min_{w} { \frac{1}{2n_{\text{samples}}} ||X W - Y||_{\text{Fro}} ^ 2 + \alpha ||W||_{21}}
+.. math::  \min_{W} { \frac{1}{2n_{\text{samples}}} ||X W - Y||_{\text{Fro}} ^ 2 + \alpha ||W||_{21}}
 
 where :math:`\text{Fro}` indicates the Frobenius norm