Inclusive growth in Pakistan the role of digitalization, education, institutional quality and technological innovation using supervised machine learning
摘要
The Government of Pakistan is accelerating digital transformation through major initiatives such as the Digital Economy Enhancement Project (DEEP), which helped boost IT exports to $3.223 billion in FY 2023–24 and expanded broadband access to over 139 million users. Programs like DigiSkills.pk have trained more than 600,000 individuals, while digital governance tools such as e-office systems and public health apps have significantly improved service delivery and efficiency. The rise of startups, adoption of AI policies, and preparations for 5G rollout reflect a shift toward a technology-driven, inclusive, and innovative economy.
PurposeTherefore, the present study investigates the effects of digitalization, education, institutional quality, and technological innovation on inclusive growth (IG) in Pakistan from 2000 to 2023.
MethodsThe research employs both regression techniques (FMOLS, DOLS, and CCR) and supervised machine learning models, including Support Vector Machine, Lasso Regression, Ridge Regression, Random Forest, Gradient Boosting, KNN, and Decision Tree.
ResultsThe finding from machine learning models indicate ICT as the most influential driver of IG, followed by technological innovation, education, and institutional quality. Regression estimates similarly confirm the positive and significant role of all four predictors in fostering IG. Among the machine learning models, Support Vector Machine demonstrated the highest predictive accuracy, while Decision Tree performed the weakest.
Policy recommendationThe study provides important policy recommendations aimed at achieving inclusive and sustainable economic development aligned with the UN Sustainable Development Goals.
Graphical abstract