In this presentation, Juan Rubio-Ramírez introduces a novel inference algorithm for structural vector autoregressions (SVARs) identified by sign restrictions. The key aspect of the algorithm is to move beyond the traditional accept–reject tradition associated with sign-identified SVARs. He demonstrates how embedding an elliptical slice sampling within a Gibbs sampler approach can deliver dramatic gains in speed and turn previously infeasible applications into feasible ones. A tractable example will illustrate the power of elliptical slice sampling for sign-identified SVARs. Finally, Juan will demonstrate the algorithm’s utility by applying it to a well-known small-SVAR model of the oil market featuring a tight identified set, and to a large-SVAR model with more than 100 sign restrictions.

In this presentation, Kaustubh addresses the imperative to enhance nowcasting models for Indian GDP in the post-pandemic era. He brings together a diverse set of high-frequency economic indicators (HFIs) into multiple factors—nominal, survey-based, labor-market, and real economic activity—to produce GDP growth nowcasts. Unlike existing models that primarily focus on aggregate estimates, he evaluates each HFI’s contribution by analyzing its impact on nowcast revisions following new data releases.

In addition, the COVID-19 pandemic introduced outliers in HFIs, distorting model parameters and reducing forecasting accuracy. To address this, Kaustubh discusses incorporating the Oxford Stringency Index and proposes a novel sigmoid-based data transformation that minimizes model sensitivity to large shocks. This approach enables models to handle unexpected events more robustly without overreacting. This methodology improves nowcasting models’ resilience to outliers, providing valuable insights during volatile periods.

The use of mixed-frequency data is now common in many applications, ranging from the analysis of high-frequency financial time series to large cross-sections of macroeconomic time series. In this presentation, David Kelley from the Federal Reserve Bank of New York discusses how state-space methods can easily facilitate both estimation and inference in these settings. He introduces a unified treatment of the state-space approach to mixed-frequency data modeling and show applications of the MFSS toolbox in MATLAB^®, which streamlines model estimation and post-processing.

In this session, Allan Wright presents a dynamic stochastic general equilibrium (DSGE) model developed by the Central Bank of The Bahamas to analyze the macroeconomic dynamics of a small open economy with distinct tourism and non-tourism sectors. The model is calibrated to a tourism-dependent economy, and features heterogeneous firms, a flexible exchange rate, and a Taylor rule. Steady-state analysis highlights that tourism accounts for 28% of output, with consumption at 1.215 and a real interest rate of 1.0101. Variance decomposition reveals that foreign interest rate shocks explain 65.68% of consumption and 69.36% of labor volatility, while tourism productivity shocks account for 32.08% of tourism output, 59.78% of aggregate output, and 91.09% of inflation fluctuations. Impulse response functions show that a tourism productivity shock raises consumption by 0.981% and aggregate output by 0.421%, while reducing inflation by 0.730%. In contrast, a foreign interest rate shock lowers consumption by 2.057% and increases inflation by 0.221%. The flexible exchange rate mitigates these effects, underscoring its stabilizing role. These findings highlight the importance of sector-specific policies and exchange rate management in tourism-driven economies. 

In this presentation, Laura Ballotta investigates the complexities underlying the calibration of option-pricing models, a crucial task for quantitative analysts in financial markets. She emphasizes the importance of algorithm selection when navigating the “valleys” of the objective-function landscape, advocating greater scrutiny of optimization routines to ensure consistent, meaningful parameter estimates. Laura also discusses how to set up adequate objective functions for joint-calibration problems.

Learn how using MATLAB^® and Databricks^® together enables seamless integration of advanced analytics into a modern data platform. This session will explore how to overcome common challenges in scaling, orchestration, and deploying MATLAB into the cloud with Databricks, and how to unlock the full potential of your data and models.

In this talk, we describe how to:

• Orchestrate to clusters, jobs, etc.
• Make database connections with JDBC/ODBC
• Call Spark^™ interactively from MATLAB
• Deploy compiled MATLAB on a cluster
• Run the MATLAB desktop on Databricks
• Deploy MATLAB on serverless clusters

Join us to learn more about how the MATLAB interface for Databricks helps you deploy, run, and scale MATLAB workflows directly in the cloud.

In this presentation, Lawrence Johnny demonstrates how MATLAB Copilot accelerates robust model development, testing, and validation—leveraging AI-powered code generation, error explanations, and automated unit tests for financial models.