Forecasting time series with encoder-decoder neural networks
This work addresses time series forecasting for applications like climate modeling, but it appears incremental as it applies an existing encoder-decoder method to this domain with theoretical analysis.
The paper tackles forecasting high-dimensional stationary time series using encoder-decoder neural networks, providing theoretical error bounds under structural and sparsity assumptions and validating the approach with temperature data.
In this paper, we consider high-dimensional stationary processes where a new observation is generated from a compressed version of past observations. The specific evolution is modeled by an encoder-decoder structure. We estimate the evolution with an encoder-decoder neural network and give upper bounds for the expected forecast error under specific structural and sparsity assumptions. The results are shown separately for conditions either on the absolutely regular mixing coefficients or the functional dependence measure of the observed process. In a quantitative simulation we discuss the behavior of the network estimator under different model assumptions. We corroborate our theory by a real data example where we consider forecasting temperature data.