Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms

Yu Jeffrey Hu; Jeroen Rombouts; Ines Wilms

Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms

Yu Jeffrey Hu, Jeroen Rombouts^*, Ines Wilms

^*Corresponding author for this work

Research output: Working paper / Preprint › Working paper

Abstract

On-demand service platforms face a challenging problem of forecasting a large collection of high-frequency regional demand data streams that exhibit instabilities. This paper develops a novel forecast framework that is fast and scalable, and automatically assesses changing environments without human intervention. We empirically test our framework on a large-scale demand data set from a leading on-demand delivery platform in Europe, and find strong performance gains from using our framework against several industry benchmarks, across all geographical regions, loss functions, and both pre- and post-Covid periods. We translate forecast gains to economic impacts for this on-demand service platform by computing financial gains and reductions in computing costs.

Original language	English
Publisher	Cornell University - arXiv
Number of pages	40
Publication status	Published - Mar 2023

Publication series

Series	arXiv.org
Number	2303.01887
ISSN	2331-8422

JEL classifications

g12 - "Asset Pricing; Trading volume; Bond Interest Rates"

Keywords

e-commerce
platform econometrics
streaming data
forecast breakdown

Access to Document

https://arxiv.org/abs/2303.01887Licence: CC BY

Cite this

@techreport{c0a14ed813cf4f4a8c5066d1d2cea004,

title = "Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms",

abstract = "On-demand service platforms face a challenging problem of forecasting a large collection of high-frequency regional demand data streams that exhibit instabilities. This paper develops a novel forecast framework that is fast and scalable, and automatically assesses changing environments without human intervention. We empirically test our framework on a large-scale demand data set from a leading on-demand delivery platform in Europe, and find strong performance gains from using our framework against several industry benchmarks, across all geographical regions, loss functions, and both pre- and post-Covid periods. We translate forecast gains to economic impacts for this on-demand service platform by computing financial gains and reductions in computing costs.",

keywords = "e-commerce, platform econometrics, streaming data, forecast breakdown",

author = "Hu, {Yu Jeffrey} and Jeroen Rombouts and Ines Wilms",

note = "Data from on-demand delivery platform Stuart",

year = "2023",

month = mar,

language = "English",

series = "arXiv.org",

number = "2303.01887",

publisher = "Cornell University - arXiv",

address = "United States",

type = "WorkingPaper",

institution = "Cornell University - arXiv",

}

TY - UNPB

T1 - Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms

AU - Hu, Yu Jeffrey

AU - Rombouts, Jeroen

AU - Wilms, Ines

N1 - Data from on-demand delivery platform Stuart

PY - 2023/3

Y1 - 2023/3

N2 - On-demand service platforms face a challenging problem of forecasting a large collection of high-frequency regional demand data streams that exhibit instabilities. This paper develops a novel forecast framework that is fast and scalable, and automatically assesses changing environments without human intervention. We empirically test our framework on a large-scale demand data set from a leading on-demand delivery platform in Europe, and find strong performance gains from using our framework against several industry benchmarks, across all geographical regions, loss functions, and both pre- and post-Covid periods. We translate forecast gains to economic impacts for this on-demand service platform by computing financial gains and reductions in computing costs.

AB - On-demand service platforms face a challenging problem of forecasting a large collection of high-frequency regional demand data streams that exhibit instabilities. This paper develops a novel forecast framework that is fast and scalable, and automatically assesses changing environments without human intervention. We empirically test our framework on a large-scale demand data set from a leading on-demand delivery platform in Europe, and find strong performance gains from using our framework against several industry benchmarks, across all geographical regions, loss functions, and both pre- and post-Covid periods. We translate forecast gains to economic impacts for this on-demand service platform by computing financial gains and reductions in computing costs.

KW - e-commerce

KW - platform econometrics

KW - streaming data

KW - forecast breakdown

M3 - Working paper

T3 - arXiv.org

BT - Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms

PB - Cornell University - arXiv

ER -