diff --git a/python/src/GeneralLinearModelAlgorithm_doc.i.in b/python/src/GeneralLinearModelAlgorithm_doc.i.in
index 33f2ce5c36c..607ce029696 100644
--- a/python/src/GeneralLinearModelAlgorithm_doc.i.in
+++ b/python/src/GeneralLinearModelAlgorithm_doc.i.in
@@ -51,8 +51,7 @@ where:
 
 with :math:`\mu_\ell(\vect{x}) = \sum_{j=1}^{n_\ell} \beta_j^\ell \varphi_j^\ell(\vect{x})` and
 :math:`\varphi_j^\ell: \Rset^\inputDim \rightarrow \Rset` the trend functions.
-
-:math:`\vect{W}` is a Gaussian process of dimension :math:`\outputDim` with zero mean and covariance function
+The Gaussian process :math:`\vect{W}` is of dimension :math:`\outputDim` with zero mean and covariance function
 :math:`C = C(\vect{\theta}, \vect{\sigma}, \mat{R}, \vect{\lambda})` (see :class:`~openturns.CovarianceModel`
 for the notations).
 
@@ -81,8 +80,7 @@ The *GeneralLinearModelAlgorithm* class estimates the coefficients :math:`\beta_
 where :math:`\vect{p}` is the vector of parameters of the covariance model (a subset of
 :math:`\vect{\theta}, \vect{\sigma}, \mat{R}, \vect{\lambda}`) that has been declared as
 *active* (by default, the full vectors :math:`\vect{\theta}` and :math:`\vect{\sigma}`).
-
-The estimation is done by maximizing the *reduced* log-likelihood of the model, see its expression below.
+The estimation is done by maximizing the *reduced* log-likelihood of the model (see its expression below).
 
 **Estimation of the parameters** :math:`\beta_j^\ell` and :math:`\vect{p}`