Bootstrap confidence bands for connectedness measures This project develops and implements bootstrap confidence bands for high-dimensional connectedness measures in economic and financial networks. The empirical object of interest is a panel of interconnected time series, decomposed into a low-dimensional common component and a high-dimensional idiosyncratic component. The methodology measures how much of the forecast error variance of each variable is attributable to common shocks and how much is attributable to idiosyncratic shocks transmitted through the system. The analysis is carried out both in the time domain and in the frequency domain, where the latter is interpreted as a finite-horizon decomposition of the same forecast error variance into low-, medium- and high-frequency components. The main computational challenge is inference. The connectedness measures are nonlinear functions of estimated factors, sparse vector autoregressive models, innovation covariance matrices, and moving-average response matrices. Analytical confidence intervals are not readily available in this high-dimensional setting. The project therefore relies on recursive bootstrap methods. Each bootstrap replication requires re-estimating the factor model, re-estimating the sparse idiosyncratic VAR, computing finite-horizon impulse-response and variance-decomposition objects, and aggregating these into time- and frequency-domain connectedness measures. Since hundreds or thousands of bootstrap replications are required for each empirical specification, the procedure is computationally intensive but highly parallelizable. The requested computing resources will be used to run these bootstrap replications for the baseline empirical analysis and for robustness checks involving alternative lag lengths, forecast horizons, frequency bands, and model specifications. The output will provide statistically interpretable confidence bands around connectedness measures that are otherwise typically reported only as point estimates. This is important for assessing whether observed changes in common, idiosyncratic, and frequency-specific connectedness are empirically meaningful rather than artifacts of estimation uncertainty. The project therefore contributes both methodologically, by making inference feasible for high-dimensional connectedness analysis, and empirically, by enabling a more credible interpretation of systemic risk and network transmission in large economic and financial systems.