Covariance structures for high-dimensional energy forecasting

Forecasts of various quantities over multiple time periods and/or spatial expanses are required to operate modern power systems. Furthermore, probabilistic forecasts necessary facilitate economic decision-making and risk management. This gives rise the challenge producing which capture dependency between variables, time, locations. The Gaussian Copula has been widely used for multivariate energy is scalable because entire structure captured by a covariance matrix; estimating this matrix in high dimensional problems remains research challenge. Here we focus on parametrising as step towards more robust estimation enable conditioning explanatory variables. We present range parametric structures strategies suitable forecasting.