# DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks

## TL;DR

DeepAR introduces a deep learning-based autoregressive recurrent network model for probabilistic time series forecasting, significantly improving accuracy over traditional methods across multiple real-world datasets.

## Contribution

The paper presents DeepAR, a novel deep learning approach for probabilistic forecasting that leverages autoregressive recurrent networks trained on numerous related time series.

## Key findings

- Achieves approximately 15% accuracy improvement over existing methods.
- Effectively models complex temporal dependencies in real-world data.
- Demonstrates robustness across diverse forecasting datasets.

## Abstract

Probabilistic forecasting, i.e. estimating the probability distribution of a time series' future given its past, is a key enabler for optimizing business processes. In retail businesses, for example, forecasting demand is crucial for having the right inventory available at the right time at the right place. In this paper we propose DeepAR, a methodology for producing accurate probabilistic forecasts, based on training an auto regressive recurrent network model on a large number of related time series. We demonstrate how by applying deep learning techniques to forecasting, one can overcome many of the challenges faced by widely-used classical approaches to the problem. We show through extensive empirical evaluation on several real-world forecasting data sets accuracy improvements of around 15% compared to state-of-the-art methods.