Abstract
In this paper, we first utilize a dynamic factor model with stochastic volatility (DFM-SV) to filter out the national factor from the local components of weekly state-level economic conditions indexes of the United States (US) over the period of April 1987 to August 2021. In the second step, we forecast the state-level factors in a panel data set-up based on the information content of corresponding state-level climate risks, as proxied by changes in temperature and its SV. The forecasting experiment depicts statistically significant evidence of out-of-sample predictability over a one-month- to one-year-ahead horizon, with stronger forecasting gains derived for states that do not believe that climate change is happening and are Republican. We also find evidence of national climate risks in accurately forecasting the national factor of economic conditions. Our analyses have important policy implications from a regional perspective.
Original language | English |
---|---|
Journal | International Review of Finance |
Volume | 24 |
Issue number | 1 |
Pages (from-to) | 154-162 |
Number of pages | 9 |
ISSN | 1369-412X |
DOIs | |
Publication status | Published - Mar 2024 |
Bibliographical note
Published online: 13 August 2023.Keywords
- Climate risks
- Dynamic factor model with stochastic volatility
- Forecasting
- Panel predictive regression
- State-level economic conditions