BISTRO: a general purpose oracle for macroeconomic time series 1

BIS Working Papers No 1337 Introducing BISTRO: a foundational model for unconditional and conditional forecasting of macroeconomic time series

An Open Benchmark for Evaluating Time Series Forecasting Methods Across Financial Markets

Holding the data fixed and evaluating roughly a dozen univariate methods without exogenous regressors, it is found that asset returns remain near unforecastable across every method family and that hybrid and machine learning methods exhibit additional forecasting power on basis spreads and bank indicators.

Working Paper Series

FinVerse: Financial Time-Series Benchmark

FinVerse is introduced, a finance-domain time-series forecasting benchmark that takes a first step toward more realistic evaluation and highlights the need for domain-aware benchmarks that evaluate models under objectives closer to real-world decision making.

The Least Trimmed Squares for time series (LTSts): extensions for policy support applications

International organizations need to monitor large amounts of economic and financial data to prevent or uncover potential problems in policies implementation. The analysis of such time series cannot ignore the potential presence of anomalies and structural changes. In this paper, we elaborate on a robust framework for t...

Let the Tree Decide: FABART. A Non-Parametric Factor Model for Nonlinear Oil Shock Transmission

The question of how oil supply news shocks transmit to real activity, financial conditions and regional labor markets is back at the center of the macroeconomic research agenda. To answer this question, we introduce the Factor Bayesian Additive Regression Tree (FABART) model, a nonlinear factor-augmented vector autoreg...