A0975
Title: Inflation target at risk: A time-varying parameter distributional regression
Authors: Yunyun Wang - Monash University (Australia) [presenting]
Abstract: A time-varying parameters distribution regression model is presented to analyze the complete conditional distribution of inflation based on the current economic state. The model incorporates random walk dynamics for the time-varying parameters. Unlike previous studies focusing on the conditional mean, the proposed model offers a comprehensive understanding of the dependence structure. A novel Markov Chain Monte Carlo algorithm is introduced that simultaneously estimates all model parameters. Additionally, the condition of monotonicity is explicitly imposed on the conditional cumulative density function, eliminating the need for a secondary procedure to ensure a monotonic estimated conditional density. The investigation centers on the risk associated with significant deviations of future inflation from the preferred range. This risk information is valuable for central banks in adjusting monetary policy to maintain stable inflation assists investors in balancing their portfolios against unexpected inflation or deflation, and helps consumers manage their spending patterns.
