reform_data reforms the data into a form that is
easier to use when calculating log-likelihood values etc.
reform_data(data, p)Returns the data reformed into a \(((n_{obs}-p+1)\times dp)\) matrix. The i:th row of the matrix contains the vector \((y_{i-1},...,y_{i-p})\)
\((dp\times 1)\), where
\(y_{i}=(y_{1i},...,y_{di})\)
\((d \times 1)\).
a matrix or class 'ts' object with d>1 columns. Each column is taken to represent
a univariate time series. Missing values are not supported.
a positive integer specifying the autoregressive order
No argument checks!
Assumes the observed data is \(y_{-p+1},...,y_0,y_1,...,y_{T}\).