Mixed-frequency data-driven forecasting the important economies' performance in a smart city: a novel RUMIDAS-SVR model
| DOI | https://doi.org/10.1108/IMDS-01-2022-0014 |
| Published date | 25 July 2022 |
| Date | 25 July 2022 |
| Pages | 2175-2198 |
| Subject Matter | Information & knowledge management,Information systems,Data management systems,Knowledge management,Knowledge sharing,Management science & operations,Supply chain management,Supply chain information systems,Logistics,Quality management/systems |
| Author | Weiqing Wang,Zengbin Zhang,Liukai Wang,Xiaobo Zhang,Zhenyu Zhang |
Mixed-frequency data-driven
forecasting the important
economies’performance in a smart
city: a novel RUMIDAS-SVR model
Weiqing Wang, Zengbin Zhang, Liukai Wang and Xiaobo Zhang
School of Economics and Management, University of Science and Technology Beijing,
Beijing, China, and
Zhenyu Zhang
School of Economics and Management, Tsinghua University, Beijing, China
Abstract
Purpose –The purpose of this study is to forecast the development performance of important economies in a
smart city using mixed-frequency data.
Design/methodolo gy/approach –This study introduces reverse unrestricted mixed-data sampling
(RUMIDAS) to support vector regression (SVR) to develop a novel RUMIDAS-SVR model. The RUMIDAS-SVR
model was estimated using a quadratic programming problem. The authors then use the novel RUMIDAS-SVR
model tofor ecast the development performance of all high-tech listed companies, an important sector of the economy
reflecting the potential and dynamism of urban economic development in Shanghai using the mixed-frequency
consumer price index (CPI) producer price index (PPI), and consumer confidence index (CCI) as predictors.
Findings –The empirical results show that the established RUMIDAS-SVR is superior to the competing
models with regard to mean absolute error (MAE) and root-mean-squared error (RMSE) and multi-source
macroeconomic predictors contribute to the development performance forecast of important economies.
Practical implications –Smart city policy makers should create a favourable macroeconomic environment, such
as controlling inflation or stabilising prices for companies within the city, and companies within the important city
economic sectors should take initiative to shoulder their responsibility to support the construction of the smart city.
Originality/value –This study contributes to smart city monitoring by proposing and developing a new
model, RUMIDAS-SVR, to help the construction of smart cities. It also empirically provides strategic insights
for smart city stakeholders.
Keywords Smart city, High-tech listed companies, Mixed-frequency data, RUMIDAS-SVR
Paper type Research paper
1. Introduction
In 2008, the idea of “Smart Earth”was first proposed by International Business Machines (IBM),
which is widely regarded as the starting point for the development of smart cities. IBM formally
introduced its smart city vision in 2010 (Harrison et al., 2010), seeking to contribute to the global
growth of cities. The smart city concept arose from the rapid population growth of cities and the
subsequent enormous expansion, which has posed great challenges to urban governance. Smart
cities can solve these development problems and are therefore actively promoted in countries
around the world. The concept of the smart city is being rapidly employed in Europe, with
Amsterdam starting to transform itself since 2009 (Camboim et al., 2019). In Barcelona (Gasc
o-
Hernandez, 2018), Russia (Sergi et al., 2019)andKorea(Kwak and Lee, 2021), the construction of
smart city has improved the urban environment, life quality and contribution to economic growth.
A novel
RUMIDAS-
SVR model
2175
The authors gratefully acknowledge financial support from the China Postdoctoral Science Foundation
(No. 2021M700380), the National Natural Science Foundation of China (71729001, 72025101), the
Humanity and Social Science Foundation of Ministry of Education of China (20YJA630024), and the
Fundamental Research Funds for the Central Universities (No. FRF-DF-20-11).
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/0263-5577.htm
Received 10 January 2022
Revised 4 April 2022
20 April 2022
Accepted 26 April 2022
Industrial Management & Data
Systems
Vol. 122 No. 10, 2022
pp. 2175-2198
© Emerald Publishing Limited
0263-5577
DOI 10.1108/IMDS-01-2022-0014
Smart cities are also developing rapidly in China. In 2009, the important report “Let
Science and Technology Lead China’s Sustainable Development”by Jiabao Wen, then
Premier of China, served as the prelude to the development of smartcities in China (Agency,
2009). In 2013, the Chinese government confirmed the first list of cities to transform into
smart cities, including Beijing and Shanghai and other first-tier cities (Wang and Ren,
2013). The full application of information and communications technology (ICT) such as 5G
and cloud computing in China has also brought great convenience to the development of
smart cities. In 2016, Chongqing municipality established three big data platforms to foster
the construction of a smart city (Guo, 2016). Despite the idea of the smart city originated in
Europe, it is now flourishing in China.
Among the many components of a smart city, a critical element is its ability to foster
economic growth (Shelton et al., 2015). In other words, the planning and development of the
urban economy is a principal component of smart city construction; green and sustainable
growth of the urban economy is the main goal of smart city construction. The city’s important
economies are the main drivers of the city’s economic development and the mainstay of
constructing a smart city, reflecting the potential and dynamism of smart city economic
development. Therefore, to better monitor the development of the city’s economy throughout
smart city building, we need to assess the performance of the city’s important economies and
make accurate forecasts of their performance prospects.
Chinese smart cities offer a multitude of solutions to monitor their economic development. The
city of Hefei, for example, uses a “point-line-surface”scheme for economic construction (Bureau,
2021). The “point-line-surface”relationship outlines a three-dimensional monitor and development
model of the city economy, where the “point”is a single company in the city, the “line”is an
industry with many companies and the “surface”is the industrial/high-tech zones with related
companies or industries, which are the important economic sectors in the city. The “line”that
makes up an urban economic development also varies from city to city, such as theheavy industry
in Shenyang, the tourism industry in Zhangjiajie or the high-tech industry in Shanghai. These
practices are very enlightening, in real-life economic monitoring, the government need to focuses
on the overall economic growth of a smart city but they can just focus on the important economics
or particular industry due to their capacity is limited. Therefore, to help the government better
monitor the economic development, we propose a new approach for monitoring urban economic
development that monitors and forecasts important economic performance in a smart city.
Nevertheless, how can we forecast and monitor the performance of these important
economies? Inspired by Piotroski (2000) and Wang et al. (2021a,b), we try to use listed companies
to symbolise the industry in a certain city as a whole because they are the major participants in
the city, such that we can use the market value of these listed firms as indicators to forecast the
performance of the smart city industry. Asgharian et al. (2013) verified that the addition of
macroeconomic variables improves the forecasting accuracy of the model. Macroeconomic
variables are usually monthly, quarterly or annual data, but the stock marketgenerates various
high-frequency data such as daily trading volume and daily price index (Narayan and Sharma,
2015). Therefore, there is a critical issue that needs to be solved, which is the mixed-frequency
data. Specifically, we need to use low-frequency data to forecast high-frequency data. The
traditional method is to choose the same-frequency data (Bams et al., 2017), but this will cause us
to lose the rich market information within the high-frequency data. Consequently, we attempt to
transform low-frequency data into high-frequency data to improve the forecasting performance
of important economies. Recently, Foroni et al. (2018) established a new reverse mixed data
sampling (RMIDAS) model that can forecast high-frequencyvariables using low-frequency data.
Subsequently, Xu et al. (2021b) used this model to forecast interest rates in the United States (US).
Additionally,the stock marketis a complex and active marketthat generates large amounts
of data with various non-linear relationships. The reverse unrestricted mixed data sampling
(RUMIDAS) model cannot handle these complex relationships effectively on its own, so we
IMDS
122,10
2176
To continue reading
Request your trial