Network centrality, diversification, and portfolio returns: Economic insights from blockchain industry

Asset allocation remains a central topic in emerging digital industry, yet traditional models often assume simple relationships and ignore complex interdependencies among assets. Recent studies have introduced network-based methods, but most fail to account for multiple forms of dependency and their economic implications. This study examines how network centrality and diversification shape portfolio performance in China's blockchain market. To obtain the complex structural relationships between firms, we construct a novel multilayer synthetic network using multidimensional correlation measures and particle swarm optimization. Empirical evidence shows that investing in central firms consistently improves returns and lowers risk. However, excessive diversification weakens portfolio efficiency, revealing a trade-off between risk reduction and return dilution. These findings are robust across various datasets, timeframes, and portfolio strategies. By connecting firm network positions to portfolio outcomes, this research advances the literature on asset allocation under structural complexity, offering new insights for investors in volatile and emerging digital markets.