Discussing the article: "Data Science and ML (Part 47): Forecasting the Market Using the DeepAR model in Python"

[MetaQuotes]

Check out the new article: Data Science and ML (Part 47): Forecasting the Market Using the DeepAR model in Python.

In this article, we will attempt to predict the market with a decent model for time series forecasting named DeepAR. A model that is a combination of deep neural networks and autoregressive properties found in models like ARIMA and Vector Autoregressive (VAR).

Time series forecasting has never been an easy task in machine learning; several techniques and models have been introduced to tackle this problem, most without definitive success. Linear and non-linear models are often not capable of this task either, despite showing glimpses of decent predictions of time series data.

To tackle time series forecasting, traders have found a resort in neural network-based models such as recurrent neural networks (RNNs).

However, RNNs are more like non-linear models and less like time series models. Those familiar with Auto Regressive Integrated Moving Average (ARIMA) and Vector AutoRegressive (AR) might have noticed this. They do require extra steps to prepare the data into windows to make the neural network aware of time series patterns, despite that they are still not programmed for seasonal patterns that traditional models for time series forecasting acknowledge.

Author: Omega J Msigwa