Partial Identification of Heteroskedastic Structural Vector Autoregressions: Theory and Bayesian Inference Staff Working Paper 2025-14 Helmut Lütkepohl, Fei Shang, Luis Uzeda, Tomasz Woźniak We consider structural vector autoregressions that are identified through stochastic volatility. Our analysis focuses on whether a particular structural shock can be identified through heteroskedasticity without imposing any sign or exclusion restrictions. Content Type(s): Staff research, Staff working papers Research Topic(s): Econometric and statistical methods, Fiscal policy JEL Code(s): C, C1, C11, C12, C3, C32, E, E6, E62
Estimating Discrete Choice Demand Models with Sparse Market-Product Shocks Staff Working Paper 2025-10 Zhentong Lu, Kenichi Shimizu We propose a novel approach to estimating consumer demand for differentiated products. We eliminate the need for instrumental variables by assuming demand shocks are sparse. Our empirical applications reveal strong evidence of sparsity in real-world datasets. Content Type(s): Staff research, Staff working papers Research Topic(s): Econometric and statistical methods, Market structure and pricing JEL Code(s): C, C1, C3, D, D1, L, L0, L00
Estimating the impacts on GDP of natural disasters in Canada Staff Analytical Note 2025-5 Tatjana Dahlhaus, Thibaut Duprey, Craig Johnston Extreme weather events contribute to increased volatility in both economic activity and prices, interfering with the assessment of the true underlying trends of the economy. With this in mind, we conduct a timely assessment of the impact of natural disasters on Canadian gross domestic product (GDP). Content Type(s): Staff research, Staff analytical notes Research Topic(s): Climate change, Domestic demand and components, Econometric and statistical methods, Fiscal policy, Regional economic developments JEL Code(s): B, B2, B23, C, C1, C13, C2, C23, E, E1, E17, E3, E37, E6, E62, H, H6
Seasonal Adjustment of Weekly Data Staff Discussion Paper 2024-17 Jeffrey Mollins, Rachit Lumb The industry standard for seasonally adjusting data, X-13ARIMA-SEATS, is not suitable for high-frequency data. We summarize and assess several of the most popular seasonal adjustment methods for weekly data given the increased availability and promise of non-traditional data at higher frequencies. Content Type(s): Staff research, Staff discussion papers Research Topic(s): Econometric and statistical methods JEL Code(s): C, C1, C4, C5, C52, C8, E, E0, E01, E2, E21
Decision Synthesis in Monetary Policy Staff Working Paper 2024-30 Tony Chernis, Gary Koop, Emily Tallman, Mike West We use Bayesian predictive decision synthesis to formalize monetary policy decision-making. We develop a case-study of monetary policy decision-making of an inflation-targeting central bank using multiple models in a manner that considers decision goals, expectations and outcomes. Content Type(s): Staff research, Staff working papers Research Topic(s): Econometric and statistical methods, Economic models, Monetary policy JEL Code(s): C, C1, C11, C3, C32, C5, C53
Non-Parametric Identification and Testing of Quantal Response Equilibrium Staff Working Paper 2024-24 Johannes Hoelzemann, Ryan Webb, Erhao Xie We show that the utility function and the error distribution are non-parametrically over-identified under Quantal Response Equilibrium (QRE). This leads to a simple test for QRE. We illustrate our method in a Monte Carlo exercise and a laboratory experiment. Content Type(s): Staff research, Staff working papers Research Topic(s): Econometric and statistical methods, Economic models JEL Code(s): C, C1, C14, C5, C57, C9, C92
Parallel Tempering for DSGE Estimation Staff Working Paper 2024-13 Joshua Brault I develop a population-based Markov chain Monte Carlo algorithm known as parallel tempering to estimate dynamic stochastic general equilibrium models. Parallel tempering approximates the posterior distribution of interest using a family of Markov chains with tempered posteriors. Content Type(s): Staff research, Staff working papers Research Topic(s): Econometric and statistical methods, Economic models JEL Code(s): C, C1, C11, C15, E, E1, E10
Predictive Density Combination Using a Tree-Based Synthesis Function Staff Working Paper 2023-61 Tony Chernis, Niko Hauzenberger, Florian Huber, Gary Koop, James Mitchell This paper studies non-parametric combinations of density forecasts. We introduce a regression tree-based approach that allows combination weights to vary on the features of the densities, time-trends or economic indicators. In two empirical applications, we show the benefits of this approach in terms of improved forecast accuracy and interpretability. Content Type(s): Staff research, Staff working papers Research Topic(s): Econometric and statistical methods JEL Code(s): C, C1, C11, C3, C32, C5, C53
Three things we learned about the Lynx payment system Staff Analytical Note 2023-14 Nikil Chande, Zhentong Lu, Hiru Rodrigo, Phoebe Tian Canada transitioned to a new wholesale payment system, Lynx, in August 2021. Lynx is based on a real-time settlement model that eliminates credit risk in the system. This model can require more liquidity; however, Lynx’s design allows Canada’s wholesale payments to settle efficiently. Content Type(s): Staff research, Staff analytical notes Research Topic(s): Financial institutions, Financial services, Financial system regulation and policies, Payment clearing and settlement systems JEL Code(s): C, C1, C10, E, E4, E42, G, G2, G28
Combining Large Numbers of Density Predictions with Bayesian Predictive Synthesis Staff Working Paper 2023-45 Tony Chernis I show how to combine large numbers of forecasts using several approaches within the framework of a Bayesian predictive synthesis. I find techniques that choose and combine a handful of forecasts, known as global-local shrinkage priors, perform best. Content Type(s): Staff research, Staff working papers Research Topic(s): Econometric and statistical methods JEL Code(s): C, C1, C11, C5, C52, C53, E, E3, E37