Scholarlink

<p>Financial inclusion is crucial for sustainable economic growth and reducing poverty. Despite efforts to promote financial inclusion, a significant portion of the global population remains unbanked, particularly in developing countries like India. The paper addresses this issue by constructing a comprehensive Financial Inclusion (FI) index and predicting future trends using advanced machine learning (ML) algorithms. The study uses data from the Global Findex Database 2023, which includes 128 variables pertaining to Indian households and employs Linear Regression, Support Vector Regression, Random Forest, and Gradient Boosting models to predict financial inclusion levels, using Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and R-squared for error measurement. All the models’ predictions are, on average, closer to the actual values. By creating a continuous FI index and applying ML techniques, this research provides a deeper understanding of financial inclusion in India, highlighting specific areas for policy intervention and enabling more targeted efforts to enhance financial access and usage. This data-driven approach can serve as a model for other developing economies aiming to improve financial inclusion and economic participation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Demystifying financial inclusion in an emerging economy: a machine learning approach to index construction and forecasting

  • 期刊论文

  • 发表日期：

    2026年2月20日

查看全文
  • R. L. Manogna,
  • Shrey Mehta,
  • Devansh Agarwal

摘要

Financial inclusion is crucial for sustainable economic growth and reducing poverty. Despite efforts to promote financial inclusion, a significant portion of the global population remains unbanked, particularly in developing countries like India. The paper addresses this issue by constructing a comprehensive Financial Inclusion (FI) index and predicting future trends using advanced machine learning (ML) algorithms. The study uses data from the Global Findex Database 2023, which includes 128 variables pertaining to Indian households and employs Linear Regression, Support Vector Regression, Random Forest, and Gradient Boosting models to predict financial inclusion levels, using Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and R-squared for error measurement. All the models’ predictions are, on average, closer to the actual values. By creating a continuous FI index and applying ML techniques, this research provides a deeper understanding of financial inclusion in India, highlighting specific areas for policy intervention and enabling more targeted efforts to enhance financial access and usage. This data-driven approach can serve as a model for other developing economies aiming to improve financial inclusion and economic participation.