This document discusses using singular spectrum analysis (SSA) to model electricity prices. SSA decomposes time series data into trend, periodic, and noise components. The author applies SSA to French electricity prices, trading volumes, and consumption data. SSA shows the price trends are better correlated with consumption data than with trading volumes. The author proposes a model where the short-range price components are a function of short-range consumption components and noise, providing a physically-based way to link prices and consumption for scenario testing and analysis.