Forecasting BDI Sea Freight Shipment Cost, VIX Investor Sentiment and MSCI Global Stock Market Indicator Indices: LSTAR-GARCH and LSTAR-APGARCH Models

Prediction of the economy in global markets is crucial importance for individuals, decisionmakers, and policies. To this end, effectiveness modeling forecasting directions such leading indicators importance. For purpose, we analyzed Baltic Dry Index (BDI), Investor Sentiment (VIX), Global Stock Market Indicator (MSCI) their distributional characteristics to proposed econometric methods. Among these, BDI an economic indicator based on shipment dry cargo costs, VIX a measure investor fear, MSCI represents emerging developed county stock market indicator. By utilizing daily data sample covering 1 November 2007–30 May 2022, BDI, VIX, indices are investigated with various methods nonlinearity, chaos, regime-switching volatility. The BDS independence test confirmed dependence nonlinearity all three series; Lyapunov exponent, Shannon, Kolmogorov entropy tests suggest that series follow chaotic processes. Smooth transition autoregressive (STAR) type favored two-regime GARCH Asymmetric Power (APGARCH) nonlinear conditional volatility models where regime changes governed by smooth logistic transitions. Nonlinear LSTAR-GARCH LSTAR-APGARCH models, addition single-regime variants, estimated evaluated in-sample out-of-sample forecasts. findings determined significant prediction forecast improvement LSTAR-APGARCH, closely followed models. Overall results confirm necessity integrating dynamics utilize as effective investors policymakers predict direction economy.