Modeling the digital maturity of national economy and verification of the macroeconomic effects of digital transformation

The study presents an assessment of the digital maturity of national economy grounded in fuzzy set theory and objective feature weighting. From a harmonized panel of information and communication technology (ICT) indicators — after series normalization with 1st-99th percentile winsorization, monotonicity control and unit harmonization — a composite maturity index is constructed, followed by stratification of countries by maturity levels. The baseline set includes the prevalence of fixed broadband access for households and businesses, the number of active mobile subscriptions per capita, the share of Internet users, the density of secure Internet servers, the share of ICT goods in total merchandise imports and the share of ICT services in services exports. Where available, positions for software and telecommunications products are added with separate accounting logic for goods and service flows. Normalized indicators are transformed into linguistic variables; triangular and trapezoidal membership functions are specified for the «low», «medium» and «high» gradations with openly parameterized vertices and plateaus suitable for scenario adjustments. The pipeline then performs fuzzification, weighted aggregation and defuzzification via the centroid method, with robustness checks against alternative membership shapes and aggregation schemes. Indicator weights are estimated using the entropy method and principal component analysis, reconciled and applied at aggregation to mitigate multicollinearity; principal component analysis loadings and component contributions are retained for interpretation. The resulting index supports ranking of economies, computation of probabilistic membership shares across maturity tiers, mapping, and clustering. External validation is carried out against the Human Development Index using correlation- regression analysis; robustness is confirmed by bootstrap estimates and heteroskedasticity-robust standard errors. The algorithm is implemented in Python with openly specified normalization settings and membership functions, which ensures reproducibility and scalability. The resulting estimates substantiate the prioritization of infrastructure, cybersecurity and digital-skills development for strategic planning, progress monitoring and international benchmarking.