Impact of Near-Time Information for Prediction on Microeconomic Balanced Time Series Data using Different Machine Learning Methods

55 Pages Posted: 10 Apr 2020

See all articles by Frederik Collin

Frederik Collin

Ulm University - Department of Mathematics and Economics

Martin Kies

Ulm University; LeverageData GmbH

Date Written: March 23, 2020

Abstract

Instead of relying solely on data of a single time series it is possible to use information of parallel, similar time series to improve prediction quality. Our data set consists of microeconomic data of daily store deposits from a large number of different stores. We analyze how prediction performance regarding a given store can be increased by using data from other stores. First we compare several machine learning methods, including Elastic Nets, Partial Least Squares, Generalized Additive Models, Random Forests, Gradient Boosting and Neural Networks using only data of a single time series. Afterwards we show that Random Forests are able to better utilize parallel time series data compared to Partial Least Squares. Using near-time data of parallel time series is highly beneficial for prediction performance. To allow a fair comparison between different machine learning methods, we present a novel hyper-parameter optimization technique using a regression tree. It enables a fast and flexible determination of optimal parameters for a given method.

Keywords: Time Series, Machine Learning, Forecasting, Nowcasting, Partial Least Squares, Random Forests, Neural Network, Hyperparameter Optimization

JEL Classification: C15, C32, C45, C53, C83

Suggested Citation

Collin, Frederik and Kies, Martin and Kies, Martin, Impact of Near-Time Information for Prediction on Microeconomic Balanced Time Series Data using Different Machine Learning Methods (March 23, 2020). Available at SSRN: https://ssrn.com/abstract=3559645 or http://dx.doi.org/10.2139/ssrn.3559645

Frederik Collin

Ulm University - Department of Mathematics and Economics ( email )

Helmholzstrasse
Ulm, D-89081
Germany

Martin Kies (Contact Author)

Ulm University ( email )

Helmholtzstr. 18
Ulm, Baden-W├╝rttemberg 89081
Germany
7315015367 (Phone)

LeverageData GmbH ( email )

Wagnerstr. 18
Ulm, 89077
Germany
+4973129879770 (Phone)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
52
Abstract Views
345
rank
504,541
PlumX Metrics