The study evaluates 13 economic indicators across six countries (India, Philippines, Thailand, France, UK, US) from 2000–2023, detecting anomalies, structural breaks, and outlier behaviour. It employs Benford’s Law, Grubbs’ Test, Chow Test, and DBSCAN on data from the Global Macro Database, using a contingency-based approach to validate anomalies through methodological convergence. Anomalies often correspond with periods of political transition or institutional volatility, emphasising the impact of political risk, institutional fragility, and data governance, especially in developing economies. The study offers a reproducible and scalable methodology for auditing official economic statistics, supported by a review of literature on economic measurement, data analytics, and political risk.

The paper explores a variety of data analytics methods—such as Benford’s Law for detecting manipulation, Markov Switching Models for economic cycle analysis, time-series anomaly detection for data integrity, and volatility analysis for stability—to assess the quality of economic data. It highlights the use of advanced techniques like Bayesian inference and resampling (e.g., bootstrap methods) alongside cross-source comparisons—such as GDP validation with satellite data, inflation checks with online pricing, and employment trends with job posting data—to identify discrepancies and enhance reliability. Thematic analysis reveals an evolving landscape of economic data assessment, progressing from probabilistic approaches to resampling techniques, with interdisciplinary integration of data analytics, econometrics, and business policy to address data quality challenges like noise, biases, and structural shifts. While these methods improve economic forecasting and policy decisions, limitations persist, including sensitivity to model assumptions, regional biases in alternative data sources, and challenges in capturing informal economies, prompting future research into advanced machine learning and hybrid validation models.

Economic loss caused by fraud has become a subject of concern for countries globally. Digital world also provides data and these can be leveraged to detect and prevent fraud while also applying forensic analytics to recover the loss. Although gathering and collating data from various sources poses a challenge, the benefits outweigh the costs. Data analytics, if implemented correctly, may detect fraud and prevent a potential economic loss. The article discusses challenges, solutions and technologies for implementing a data-driven approach.

Goods and service tax (GST) is a value-added tax which is levied on goods and services sold and consumed domestically within a country. Although GST is paid by customers it is remitted to the government by the businesses selling the goods and services. The implementation of GST in India is a relatively new development that has impacted on fiscal transfers. The Fifteenth Finance Commission of India is currently deliberating on its terms of reference to determine fiscal transfers from the centre to state governments for the period 2020/1 to 2024/5. The GST Network (GSTN) has been established to provide information technology infrastructure to taxpayers, central and state governments, dealers and all stakeholders. Evidently, there are substantial opportunities to leverage data emanating from GSTN. In such a context, the role of data analytics becomes prominent in monitoring tax administration, mitigating tax evasion, leveraging digitisation and designing fiscal federal policy. The implications presented in this article are relevant to any country having a federal structure that has implemented GST in some form or another.