Research Dataset

<pre>This study investigates the predictive accuracy and potential biases of market-based inflation expectations, focusing on zero-coupon inflation swaps and breakeven inflation rates. By employing a regime-dependent framework grounded in Markov-switching autoregressive models, the analysis distinguishes between high- and low-volatility periods to evaluate the performance of these instruments. The findings reveal that short-term instruments (e.g., 12-month swaps) provide near-unbiased forecasts, whereas longer-term instruments exhibit systematic biases, especially during volatile periods, reflecting evolving risk premia and liquidity constraints. Structural break analysis uncovers significant shifts in the relationship between market-based inflation expectations and realized inflation, particularly during disruptive events such as the COVID-19 pandemic and the 2021 energy crisis. Additionally, logistic regressions highlight the VIX as a robust predictor of severe forecast errors, underscoring the amplifying effect of financial market turbulence on inflation expectations, while heightened credit market stress (TED Spread) appears to mitigate extreme forecast deviations, potentially reflecting cautious investor behavior or stabilizing central bank interventions. These results underscore the importance of regime-sensitive forecasting approaches and the integration of systemic risk indicators to improve the reliability of market-based inflation expectations in volatile macro-financial environments.<br></pre>