Signal Processing - May 2017 - 14

Xiao-Ping (Steven) Zhang
and Fang Wang

Concepts, framework, and big data applications

conomic data and financial markets are intriguing to
researchers working on data and quantitative models. With
rapid growth of and increasing access to data in digital
form, finance, economics, and marketing data are poised to
become one of the most important and tangible big data applications, owing not only to the relatively clean organization and
structure of the data but also to clear application objectives and
market demands. However, data-related economic studies often
have different viewpoints from signal processing (SP). Also,
many fundamental economics and business problems have been
well formulated and studied in both theory and practice. The
knowledge of foundational finance and economic theories will
help SP and data researchers avoid reinventing the wheel and
develop meaningful and useful research in these areas.
This tutorial article intends to introduce the mainstream
foundational concepts and framework in finance, economics,
and marketing research, elaborate on the relationships between
the traditional economic research paradigm and SP methodology, and help SP researchers identify relevant research directions. The article aims to present a refreshing SP perspective of
finance, economics, and marketing research as well as in-depth
examples on SP applications in these fields. We hope to empower
SP researchers to broaden their knowledge beyond their current

Digital Object Identifier 10.1109/MSP.2017.2663138
Date of publication: 26 April 2017

©istockphoto.com/Enis Aksoy

Signal Processing
for Finance, Economics,
and Marketing

areas of expertise and quickly grasp the right formulations of the
research questions and related evaluation criteria in these fields.

Introduction
For SP researchers, comparing an economic system, either a
financial market or a consumer market, with a familiar physical
input-output system with an impulse-response relationship is
intriguing. However, it is important to understand the different
and unique perspectives of an economic system. An economic
system is an open-loop system in which humans are active participants. System models to be used in analysis are not known a
priori and for certain. In economic studies, researchers always
work with assumptions or hypotheses that cannot be verified
using physical or natural laws. In addition, it is difficult to perform controlled experiments as is usually done in SP systems.
Increasing amounts of digital market data are readily
available, due to the rapid growth of electronic trading along
with the development of broadband Internet. Electronic trading platforms can record every bid and ask as well as every
high-frequency transaction. Indeed, finance, economics, and
marketing data are poised to become one of the most important and tangible big data applications not only because of the
relative clean organization and structure of the data but also
because of clear application objectives and market demands.
For example, owing to its familiar appearance, financial
time series has become a major interest for SP researchers,
as is evident by recent special issues [1]-[3] on related topics.

IEEE SIgnal
Signal ProcESSIng
Processing MagazInE
Magazine

May 2017

http://www.istockphoto.com/Enis

Table of Contents for the Digital Edition of Signal Processing - May 2017