Grammar mistake

chapters/04/04x-examples.tex

@@ -288,7 +288,7 @@ \subsection{Random effects models}
 Then, each of these $f^j$ has a slope and intercept for which we can estimate from the fitted regression lines $\hat f^j(x_{ij})$, $i=1,\dots,n_j$.
 This would give us the posterior mean estimates of the random intercepts and slopes.
 In order to obtain these intercepts and slopes, we simply run a best fit line through the I-prior estimated \code{conc} values.
-Furthermore, as $\sigma_0^2$ and $\sigma_1^2$ represent measures of group variability for the intercepts and slopes respectively, we can also calculate this manually for the 10 intercepts and slopes of the fitted I-prior model.
+Furthermore, as $\sigma_0^2$ and $\sigma_1^2$ represent measures of group variability for the intercepts and slopes respectively, we can also calculate these manually for the 10 intercepts and slopes of the fitted I-prior model.
 In the same spirit, $\rho_{01} = \sigma_{01} / (\sigma_0 \sigma_1)$, which is the correlation between the random intercept and slope, can also be calculated.
 
 \cref{fig:IGF.plot.beta} illustrates the differences in the estimates for the random coefficients, while  \cref{tab:igf} illustrates the differences in the estimates for the covariance matrix.

chapters/R/04x-examples.Rnw

@@ -130,7 +130,7 @@ Denote by $f^j$ the individual linear regression lines for each of the $j=1,\dot
 Then, each of these $f^j$ has a slope and intercept for which we can estimate from the fitted regression lines $\hat f^j(x_{ij})$, $i=1,\dots,n_j$.
 This would give us the posterior mean estimates of the random intercepts and slopes.
 In order to obtain these intercepts and slopes, we simply run a best fit line through the I-prior estimated \code{conc} values.
-Furthermore, as $\sigma_0^2$ and $\sigma_1^2$ represent measures of group variability for the intercepts and slopes respectively, we can also calculate this manually for the 10 intercepts and slopes of the fitted I-prior model.
+Furthermore, as $\sigma_0^2$ and $\sigma_1^2$ represent measures of group variability for the intercepts and slopes respectively, we can also calculate these manually for the 10 intercepts and slopes of the fitted I-prior model.
 In the same spirit, $\rho_{01} = \sigma_{01} / (\sigma_0 \sigma_1)$, which is the correlation between the random intercept and slope, can also be calculated.
 
 \cref{fig:IGF.plot.beta} illustrates the differences in the estimates for the random coefficients, while  \cref{tab:igf} illustrates the differences in the estimates for the covariance matrix.