Hostname: page-component-7b7567c8c-tnkh4 Total loading time: 0 Render date: 2026-11-17T10:13:48.254Z Has data issue: false hasContentIssue false

Public sector innovation outcome-driven sustainable development in Bangladesh: applying the dynamic autoregressive distributed lag simulations and Kernel-based regularised least square machine learning algorithm approaches

Published online by Cambridge University Press:  12 December 2022

Md. Monirul Islam*
Affiliation:
Graduate School of Economics and Management (GSEM), Ural Federal University (UrFU), Yekaterinburg, Russian FederationGovernance and Public Policy (GPP), Bangladesh Institute of Governance and Management (BIGM), University of Dhaka (Affiliated), Dhaka 1207, Bangladesh
Mohammad Tareque
Affiliation:
Bangladesh Institute of Governance and Management (BIGM), University of Dhaka (Affiliated), Dhaka 1207, Bangladesh
*
*Corresponding author. E-mails: monirul.islam@bigm.edu.bd, mdmonirul.islam@urfu.ru
Rights & Permissions [Opens in a new window]

Abstract

This research investigates the role of public sector innovation outcomes, e.g. trademark innovation, information and communication technology (ICT), renewable energy, and governance, in the sustainable development of Bangladesh during 1980–2019. Utilising the dynamic autoregressive distributed lag (DARDL) simulation approach, this study divulges a favourable long-term influencing profile of public sector innovation outcomes, i.e. trademark innovation, ICT, and renewable energy on sustainable development, while governance has a heterogeneous impact. Besides, the findings from the DARDL simulations area plots display 10% counterfactual shocks to the public sector innovation outcomes on sustainable development. Furthermore, the Kernel-based regularised least square machine learning algorithm approach used in the study examines the marginal effects of the public sector innovation outcomes on sustainable development for robust findings. Therefore, the policy suggestions are solely concerned with the public sector’s adoption of more innovation dynamics through appropriate policy formulation.

Information

Type
Research Article
Copyright
© The Author(s), 2022. Published by Cambridge University Press
Figure 0

Figure 1.Public sector innovation-driven sustainable development.Sources: Author’s modified framework emanated from the ideas of Briggs (2007); Nählinder and Eriksson (2019) and Arundel et al. (2019).

Figure 1

Table 1.Descriptions of variables

Figure 2

Table 2.Summary statistics of dataset

Figure 3

Table 3.Unit-root analysis

Figure 4

Table 4.ARDL bounds test with F-statistic

Figure 5

Table 5.The dynamic ARDL simulations analysis

Figure 6

Table 6.Model diagnostic statistics

Figure 7

Figure 2Hypothesised empirical framework for public sector innovation-driven sustainable development.

Figure 8

Figure 310% (±) shock to trademark innovation (LnTDK) on sustainable development (LnSD) produced using the dynamic ARDL simulations area plot. The black line indicates the average predicted value and the blue area from darkest to lightest depicts the 75%, 90%, and 95% confidence intervals.

Figure 9

Figure 410% (±) shock to information and communication technology (LnICT) on sustainable development (LnSD) produced using the dynamic ARDL simulations area plot. The black line indicates the average predicted value and the blue area from darkest to lightest depicts the 75%, 90%, and 95% confidence intervals.

Figure 10

Figure 510% (±) shock to renewable energy (LnREN) on sustainable development (LnSD) produced using the dynamic ARDL simulations area plot. The black line indicates the average predicted value, and the blue area from darkest to lightest depicts the 75%, 90%, and 95% confidence intervals.

Figure 11

Figure 610% (±) shock to governance (GOV) on sustainable development (LnSD) produced using the dynamic ARDL simulations area plot. The black line indicates the average predicted value, and the blue area from darkest to lightest depicts the 75%, 90%, and 95% confidence intervals.

Figure 12

Table 7.KRLS using pointwise derivatives

Figure 13

Figure 7Pointwise marginal effect of trademark innovation.

Figure 14

Figure 8Pointwise marginal effect of ICT.

Figure 15

Figure 9Pointwise marginal effect of renewable energy.

Figure 16

Figure 10Pointwise marginal effect of governance.

Figure 17

Figure A1.CUSUM and CUSUM squares tests.

Supplementary material: Link

Islam and Tareque Dataset

Link