Probabilistic Time Series Forecasting with Implicit Quantile Networks
This work addresses probabilistic forecasting for time series applications, representing an incremental improvement through hybrid methodology.
The authors tackled probabilistic time series forecasting by combining an autoregressive recurrent neural network with Implicit Quantile Networks to model temporal distributions, achieving favorable results in point-wise prediction accuracy and distribution estimation compared to other models on real and simulated data.
Here, we propose a general method for probabilistic time series forecasting. We combine an autoregressive recurrent neural network to model temporal dynamics with Implicit Quantile Networks to learn a large class of distributions over a time-series target. When compared to other probabilistic neural forecasting models on real- and simulated data, our approach is favorable in terms of point-wise prediction accuracy as well as on estimating the underlying temporal distribution.