deletions | additions       diff --git a/materials_and_methods.tex b/materials_and_methods.tex  index c4a1073..3fe129b 100644  --- a/materials_and_methods.tex  +++ b/materials_and_methods.tex 
...
Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text Text. \cite{Neurobot2013} might want to know about text text text text  \subsection{Fitted Empirical Orthogonal Functions}  Fitted EOF analysis explores covariance structure of fitted values. Fitted values are obtained by multivariate regression of response matrix (temperature) on independent variables (ENSO, Volcanic Aerosols). EOF analysis is carried out on fitted covariance matrix to identify fitted EOFs and fitted PCs. This approach differs from cook (2007) in a way that instead of inverse regression (\begin{math}E(\mathbf{X}/\mathbf{y})\end{math}), $(\begin{equation}E(\mathbf{X}/\mathbf{y})\end{equation})$,  forward regression (\begin{equation}E(\mathbf{y}/\mathbf{X})\end{equation}) $(\begin{equation}E(\mathbf{y}/\mathbf{X})\end{equation})$  is used to obtain fitted EOFs. This is due to the fact that dimensionality of response (\begin{math}\mathbf{y}\end{math}) $(\begin{equation}\mathbf{y}\end{equation})$  instead of predictors (\begin{math}\mathbf{X}\end{math}) $(\begin{equation}\mathbf{X}\end{equation})$  is to be reduced to map leading spatial patterns accounting for the relationship of response and predictors. Let (\begin{math}\mathbf{Y}\end{math}) $(\begin{equation}\mathbf{Y}\end{equation})$  be the ($n \times p$) matrix of response variables and (\begin{math}\mathbf{X}\end{math}) $(\begin{equation}\mathbf{X}\end{equation})$  be the ($n \times q$) vector of independent variables. Then, multivariate regression model of \begin{math}\mathbf{Y}\end{math} $\begin{equation}\mathbf{Y}\end{equation}$  on \begin{math}\mathbf{X}\end{math} $\begin{equation}\mathbf{X}\end{equation}$  is,\\ \begin{equation}  \mathbf{Y} = \mathbf{X B} + \mathbf{E}  \end{equation}\\  Error term \begin{math}\mathbf{E}\end{math} $\begin{equation}\mathbf{E}\end{equation}$  is ($n \times p$) zero mean noise matrix where each row (ith sample) is assumed to follow multivariate normal distribution (\begin{math}N_{q}(\mathbf{0},\mathbf{\Sigma})\end{math}) $(\begin{equation}N_{q}(\mathbf{0},\mathbf{\Sigma})\end{equation})$  and \begin{math}\mathbf{B}\end{math} $\begin{equation}\mathbf{B}\end{equation}$  is ($q \times p$) matrix of regression co - efficients. The ordinary least square estimator of \begin{math}\mathbf{B}\end{math} $\begin{equation}\mathbf{B}\end{equation}$  is, \begin{equation}  \mathbf{B} = \mathbf{X^{t}X}^{-1} \mathbf{X^{t}Y}  \end{equation}\\  This is equivalent to performing p-univariate regressions. Principal component analysis is conducted on covariance matrix \begin{math}\mathbf{\Sigma_{f}}\end{math} $\begin{equation}\mathbf{\Sigma_{f}}\end{equation}$  of fitted values (\begin{math}\mathbf{\hat{Y}}\end{math}). $(\begin{equation}\mathbf{\hat{Y}}\end{equation})$.  This is done by calculating eigenvalue-eigenvector pairs \begin{math}(\lambda_{i}, \mathbf{e}_{i})\end{math} $\begin{equation}(\lambda_{i}, \mathbf{e}_{i})\end{equation}$  associated with \begin{math}\mathbf{\Sigma_{f}}\end{math}. $\begin{equation}\mathbf{\Sigma_{f}}\end{equation}$.  Then, \begin{math}\mathbf{e}_{i}\end{math} $\begin{equation}\mathbf{e}_{i}\end{equation}$  is called ith fitted EOF and \begin{math}\mathbf{e}_{i}^{t} $\begin{equation}\mathbf{e}_{i}^{t}  \mathbf{Y} \end{math} \end{equation}$  is the ith fitted PC $(i= 1,2,….,p)$. \begin{table}  \caption{Maximum size of the Manuscript\label{Tab:01}}