Can We Predict Statistical Capacity? A Study on Predicting Statistical Performance Indicators

Nadar, Binu Jemima Christopher (2025) Can We Predict Statistical Capacity? A Study on Predicting Statistical Performance Indicators. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (806kB) | Preview

Abstract

This study explores the Statistical Performance Indicators (StatPI) developed by the World Bank to evaluate national statistical capacity and support monitoring of Sustainable Development Goals (SDGs). The research addresses two core objectives: identifying data anomalies within the StatPI dataset and assessing the feasibility of predictive modeling to estimate StatPI indicator scores. Through a Data Gap Analysis (DGA), we uncovered inconsistencies between StatPI scores and data availability in the World Development Indicators (WDI), often due to differing source databases. While some discrepancies were explainable, few raised concerns about the reliability and consistency of StatPI scores.

The second part of the study applied predictive modelling, specifically Random Forest regression to assess whether SPI scores could be predicted based on available features. While models using limited binary indicators showed high accuracy, they posed a risk of overfitting. A more complex model using a broader feature set demonstrated more reliable potential, despite a slightly lower performance. Our findings suggest that while predictive analysis for StatPI is currently underutilized, it holds promise; particularly for imputing missing values or signalling declining data capacity.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Hamill, David UNSPECIFIED
Subjects:	H Social Sciences > HA Statistics Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	26 Aug 2026 09:35
Last Modified:	26 Aug 2026 09:35
URI:	https://norma.ncirl.ie/id/eprint/9649

Actions (login required)

View Item