Working Papers
"Economic Policy Uncertainty and Forecast Bias in the Survey of Professional Forecasters"
This paper analyzes the properties of forecast bias in the Survey of Professional Forecasters in relation to economic policy uncertainty. Employing the quarterly forecast bias of 14 key macroeconomic variables and 12 measures of policy uncertainty from 1985 to 2020, we demonstrate that most real activity variables have significant negative responses to economic policy uncertainty. On the other hand, there is a substantial degree of sluggishness in the corresponding forecasts, generating long-lasting forecast bias. In other words, our results show that inattentive forecasters cause SPF forecast bias using both static and dynamic frameworks.
"Unraveling the Impact of Economic Policy Uncertainty on Macroeconomic Forecasts" with Hyeongwoo Kim and Ying Lin
We explore the dynamic impact of sudden changes in the historical Economic Policy Uncertainty (EPU) index on private sector forecasts for six key macroeconomic variables in the United States from 1970 to 2014. Using a combination of univariate and multivariate structural break tests, we identify endogenously chosen break dates that capture both qualitative and quantitative shifts in forecasters' responses to EPU shocks. Interestingly, the qualitative responses of macroeconomic variables to such shocks remain consistent. Our analysis unveils two notable breaks within the sample period: the first occurred in the 1980s, followed by a second break in late 2007. By examining impulse responses to EPU shocks, we detect abrupt changes in the reactions of macroeconomic variables since the early 1980s.
Work in Progress
"Factor-augmented Forecast Models for Highly Persistent Bounded Variables"
We compare the predictability of an array of benchmark models with data dimensionality reduction approaches. Our main focus is on percentage variables during 1960-2020. We propose factor-based out of sample forecasting models to estimate latent common factors employing different approaches that include the Principal Component Analysis (PC), Partial Least Squares (PLS), and the LASSO. The common factors extracted for a large panel of 216 quarterly frequency US macroeconomic time series data. We consider both stationary (Autoregressive) and nonstationary (Random Walk) models that are augmented by latent factors to formulate out of- sample forecasts of the interest rate variables.
"FOMC and Energy Market Connection"
Other Works:
"ChatGPT in Teaching and Learning: A Systematic Review" (2024, with Duha Ali, Yasin Fatemi, Mohsen Mikfar, Jude Ugwuoke and Haneen Ali) Education Sciences.
The increasing use of artificial intelligence (AI) in education has raised questions about the implications of ChatGPT for teaching and learning. A systematic literature review was conducted to answer these questions, analyzing 112 scholarly articles to identify the potential benefits and challenges related to ChatGPT use in educational settings. The selection process was thorough to ensure a comprehensive analysis of the current academic discourse on AI tools in education. Our research sheds light on the significant impact of ChatGPT on improving student engagement and accessibility and the critical issues that need to be considered, including concerns about the quality and bias of generated responses, the risk of plagiarism, and the authenticity of educational content. The study aims to summarize the utilizations of ChatGPT in teaching and learning by addressing the identified benefits and challenges through targeted strategies. The authors outlined some recommendations that will ensure that the integration of ChatGPT into educational frameworks enhances learning outcomes while safeguarding academic standards.