diff --git a/docs/source/nnls_covariance.rst b/docs/source/nnls_covariance.rst
index ece6f58..f95d246 100644
--- a/docs/source/nnls_covariance.rst
+++ b/docs/source/nnls_covariance.rst
@@ -50,7 +50,7 @@
 covariance of :math:`y`, then the maximum likelihood problem to be
 solved is
 
-o.. math:: x^* = \arg \min_x f'(x) S^{-1} f(x)
+.. math:: x^* = \arg \min_x f'(x) S^{-1} f(x)
 
 and the corresponding covariance estimate of :math:`x^*` is given by
 
diff --git a/docs/source/nnls_modeling.rst b/docs/source/nnls_modeling.rst
index 8b84066..be87149 100644
--- a/docs/source/nnls_modeling.rst
+++ b/docs/source/nnls_modeling.rst
@@ -1331,7 +1331,7 @@
    element.
 2. For all :math:`y` on the manifold, :math:`\boxplus(x,
    \boxminus(y,x)) = y`. This ensures that any :math:`y` can be
-   reached from math:`x`.
+   reached from :math:`x`.
 3. For all :math:`\Delta`, :math:`\boxminus(\boxplus(x, \Delta), x) =
    \Delta`. This ensures that :math:`\boxplus` is an injective
    (one-to-one) map.