currency, and stock markets using a mutual information measure based on generalized entropies and Bayesian network analysis. The analysis is grounded in daily data on the total stock index, the free-market exchange rate (US dollar), and the price per gram of 18-carat gold over the period 2019–2025. Marginal and joint probability mass functions of the variables were estimated through optimal classification schemes, followed by the estimation of mutual information coefficients derived from Rényi entropy. The empirical findings reveal significant mutual information among the stock index, gold, and currency markets. According to the Bayesian network analysis, the magnitude of dependence-i.e., the mutual information content across markets- remained consistent across all causal structures. The strongest information linkage was observed between the currency and gold markets, while the weakest was between the stock and gold markets. A comparison of the two methodologies indicates that both approaches yield nearly identical rankings of the markets' mutual information content. These findings provide evidence in support of asset integration theory and confirm the existence of informational interdependence and spillovers among the gold, currency, and stock markets, which are inconsistent with full informational efficiency-particularly in its semi-strong form. Furthermore, the combined application of entropy-based and Bayesian network methods, grounded in the empirical probabilities of risky asset values, offers a novel and more transparent analytical tool for measuring information content across capital markets-an approach.