Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Posts
portfolio
publications
Identification and Estimation of Structural Factor Models with External Instruments
Working Paper, 2025
We develop a new estimator for the impulse response functions structural factor models with the use of external instruments. The proposed estimator is able to allow for the number of primitive shocks to be less than the number of static factors, and via the use of a minimum distance framework, jointly utilize multiple instruments. The minimum distance framework naturally leads to an overidentification test for the joint validity of instruments, and an auto- matic moment selection procedure to select the correct instruments. Simulation results show the improvement in the estimation accuracy of impulse response functions when more than one valid instrument is used, as well as the size and consistency of the overidentification test and automatic moment selection procedures. We apply the proposed methodology to estimate the effects of a monetary policy shock using a U.S. macroeconomic dataset with the use of popular monetary policy instruments. The results show these monetary policy instruments are all jointly valid, and that their joint use can result in more accurate and reasonable estimates of the impulse response functions.
Identification and Estimation of Structural Factor Models with External Instruments
Working Paper, 2025
We develop a new estimator for the impulse response functions structural factor models with the use of external instruments. The proposed estimator is able to allow for the number of primitive shocks to be less than the number of static factors, and via the use of a minimum distance framework, jointly utilize multiple instruments. The minimum distance framework naturally leads to an overidentification test for the joint validity of instruments, and an auto- matic moment selection procedure to select the correct instruments. Simulation results show the improvement in the estimation accuracy of impulse response functions when more than one valid instrument is used, as well as the size and consistency of the overidentification test and automatic moment selection procedures. We apply the proposed methodology to estimate the effects of a monetary policy shock using a U.S. macroeconomic dataset with the use of popular monetary policy instruments. The results show these monetary policy instruments are all jointly valid, and that their joint use can result in more accurate and reasonable estimates of the impulse response functions.
Factor-Augmented Forecasting Subject to Structural Breaks in the Factor Structure
Working Paper, 2026
This paper investigates the impact of structural breaks in the factor structure on factor-augmented forecasting. We decompose the break in the factor loading matrix into rotational and shift components. To effectively utilize pre-break data while maintaining robustness against shift breaks, we propose a novel ``rotated factor’ estimator that minimizes the $L_2$ distance between pre- and post-break loading matrices. We show that the rotated factor estimator achieves the standard convergence rate even under large shift breaks, provided the rotational change is small. Furthermore, we derive the out-of-sample bias–variance trade-offs for competing factor estimators, demonstrating that the rotated factors weakly dominate the conventional full-sample factors by purging shift-break bias, while also outperforming split-sample methods through variance reduction under small rotational changes. To leverage the respective advantages of each estimator in a data-driven way, we introduce a leave-$h$-out cross-validation averaging procedure. Simulations and an empirical application to forecasting post-COVID U.S. macroeconomic data demonstrate that our approach leads to improved forecasting performance and underscores the importance of incorporating structural breaks into factor-augmented models.
talks
teaching
Teaching 2020
Courses, Monash University, Department of Econometrics and Business Statistics, 2020
Introductory and masters level courses in econometrics, covering topics such as linear regression, binary outcome models, statistical inference, computational statistics, data visualisation and analytics
Teaching 2021
Courses, Monash University, Department of Econometrics and Business Statistics, 2021
Introductory and masters level courses in econometrics, covering topics such as linear regression, binary outcome models, statistical inference, and computational statistics
Teaching 2022
Courses, Monash University, Department of Econometrics and Business Statistics, 2022
Introductory and masters level courses in econometrics, covering topics such as linear regression, binary outcome models, statistical inference, computational statistics, data visualisation and analytics
Teaching 2023
Courses, Monash University, Department of Econometrics and Business Statistics, 2023
Introductory and masters level courses in econometrics, covering topics such as linear regression, binary outcome models, statistical inference, computational statistics, data visualisation and analytics. For my teaching in ETC3410 Applied Econometrics I was awarded a Purple Letter of Commendation for being on the teaching team of one of the top performing units in terms of student evaluations.
Teaching 2024
Courses, Monash University, Department of Econometrics and Business Statistics, 2024
Introductory and masters level courses in econometrics, covering topics such as linear regression, binary outcome models, statistical inference, computational statistics, data visualisation and analytics