corrected some pages

Several small corrections were done to several proofs and definitions.

Author: JoramSoch

P/bvn-mle.md

@@ -46,7 +46,7 @@ $$ \label{eq:mvn-data}
 y_i = \left[ \begin{matrix} y_{i1} \\ \vdots \\ y_{ip} \end{matrix} \right] \sim \mathcal{N}\left( \mu, \Sigma \right), \quad i = 1, \ldots, n
 $$
 
-for which [maximum likelihood estimates are given by](/D/mvn-mle)
+for which [maximum likelihood estimates are given by](/P/mvn-mle)
 
 $$ \label{eq:mvn-mle}
 \begin{split}

P/fren-dec.md

@@ -110,7 +110,7 @@ $$ \label{eq:vb-fe2-qed}
 \end{split}
 $$
 
-where the first term can be seen as an [accuracy term](/D/lme-anc) (= posterior expected log-likelihood) and the second term can be seen as a [complexity penalty](/D/lme-anc) (= divergence of posterior from prior distribution);
+where the first term can be seen as an [accuracy term](/P/lme-anc) (= posterior expected log-likelihood) and the second term can be seen as a [complexity penalty](/P/lme-anc) (= divergence of posterior from prior distribution);
 
 3) the sum of expected joint log-likelihood and approximate posterior entropy
 

P/matn-trans.md

@@ -34,7 +34,7 @@ X^\mathrm{T} \sim \mathcal{MN}(M^\mathrm{T}, V, U) \; .
 $$
 
 
-**Proof:** For a [random vector](/P/rvec) $X \in \mathbb{R}^n$ with [probability density function](/D/pdf) $f_X(x)$, the [probability density function of the invertible function](/P/pdf-invfct) $Y = g(X)$ is
+**Proof:** For a [random vector](/D/rvec) $X \in \mathbb{R}^n$ with [probability density function](/D/pdf) $f_X(x)$, the [probability density function of the invertible function](/P/pdf-invfct) $Y = g(X)$ is
 
 $$ \label{eq:pdf-invfct}
 f_Y(y) = \left\{

P/mlr-rssdist.md

@@ -133,7 +133,7 @@ $$ \label{eq:rss-dist}
 \end{split}
 $$
 
-Because a [non-central chi-squared distribution with non-centrality parameter of zero reduces to the central chi-squared distribution](/P/chi2-ncchi2), we obtain our final result:
+Because a [non-central chi-squared distribution with non-centrality parameter of zero reduces to the central chi-squared distribution](/P/ncchi2-chi2), we obtain our final result:
 
 $$ \label{eq:rss-dist-qed}
 \frac{\hat{\varepsilon}^\mathrm{T} \hat{\varepsilon}}{\sigma^2} \sim \chi^2(n-p) \; .

P/mvn-maxent.md

@@ -64,7 +64,7 @@ $$ \label{eq:int-fg-s2}
 \int_{\mathcal{X}} f(x) \log g(x) \, \mathrm{d}x = - \frac{n}{2} \log (2 \pi) - \frac{1}{2} \log |\Sigma| - \frac{1}{2} \left\langle (x-\mu)^\mathrm{T} \Sigma^{-1} (x-\mu) \right\rangle_{f(x)} \; .
 $$
 
-Using the [expectation of a trace](/D/mean-tr) and the [definition of the covariance matrix](/D/covmat), the second term can be developed as follows:
+Using the [expectation of a trace](/P/mean-tr) and the [definition of the covariance matrix](/D/covmat), the second term can be developed as follows:
 
 $$ \label{eq:int-fg-s3}
 \begin{split}

P/rsq-mmm.md

@@ -46,7 +46,7 @@ $$ \label{eq:rsq-dist}
 R^2 \sim \mathrm{Bet}\left( \frac{p-1}{2}, \frac{n-p}{2} \right) \; .
 $$
 
-Using [mean](/D/beta-mean), [median](/D/beta-med) and [mode](/D/beta-mode) of the [beta distribution](/D/beta)
+Using [mean](/P/beta-mean), [median](/P/beta-med) and [mode](/P/beta-mode) of the [beta distribution](/D/beta)
 
 $$ \label{eq:beta-mmm}
 \begin{split}

P/rsq-var.md

@@ -40,7 +40,7 @@ $$ \label{eq:rsq-dist}
 R^2 \sim \mathrm{Bet}\left( \frac{p-1}{2}, \frac{n-p}{2} \right) \; .
 $$
 
-Using the [variance of the beta distribution](/P/var-beta)
+Using the [variance of the beta distribution](/P/beta-var)
 
 $$ \label{eq:beta-var}
 X \sim \mathrm{Bet}(\alpha, \beta) \\