ABSTRACT
This study introduces a first set of uncertainty indexes for Uruguay (a newspaper-based index and a composite index-based) to analyze how economic uncertainty impacts domestic variables in a small and open economy such as Uruguay, which is exposed to international, regional, and local uncertainty. The analysis covers approximately 15 years and uses the vector autoregressive methodological framework. The main findings suggest that economic uncertainty significantly impacts the real economy and does not impact the nominal variables. These findings which differentiate from other results found in the empirical literature, can be associated with the stability of the Uruguayan economy and the strong institutions, which may help mitigate external shocks. To assess the capability of the proposed uncertainty model to predict macroeconomic variables, we evaluate its predictive performance within the last major uncertainty shock due to the COVID-19 pandemic. © 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG.
ABSTRACT
This study investigates the predictive ability of categorical economic-policy uncertainty (EPU) indices for stock-market returns. The results indicate that some categorical EPU indices have superior predictive ability for stock returns and even achieve higher realized utility than the original EPU index and popular predictors. Furthermore, the diffusion indices based on EPU categories, especially those that use partial least squares (PLS) to extract the principal components, more effectively use the forecast information contained in categorical EPU indices, resulting in improved forecast performance, including reduced forecast errors and increased economic value for investors. In addition, the categorical EPU indices show superior forecasting performance during economic-expansion, the China-US trade-war, and COVID-19 pandemic periods.
ABSTRACT
The quantification of economic uncertainty is key to the prediction of macroeconomic variables, such as gross domestic product (GDP), and is particularly crucial in regard to real-time or short-time prediction methodologies, such as nowcasting, where a large amount of time series data is required. Most of the data comes from official agency statistics and non-public institutions, but these sources are susceptible to lack of information due to major disruptive events, such as the COVID-19 pandemic. Because of this, it is very common nowadays to use non-traditional data from different sources. The economic policy uncertainty (EPU) index is the indicator most frequently used to quantify uncertainty and is based on topic modeling of newspapers. In this paper, we propose a methodology to estimate the EPU index that incorporates a fast and efficient method for topic modeling of digital news based on semantic clustering with word embeddings, allowing us to update the index in real time, which is something that other studies have failed to manage. We show that our proposal enables us to update the index and significantly reduce the time required for new document assignation into topics. © 2022 Elsevier Ltd
ABSTRACT
The quantification of economic uncertainty is key to the prediction of macroeconomic variables, such as gross domestic product (GDP), and is particularly crucial in regard to real-time or short-time prediction methodologies, such as nowcasting, where a large amount of time series data is required. Most of the data comes from official agency statistics and non-public institutions, but these sources are susceptible to lack of information due to major disruptive events, such as the COVID-19 pandemic. Because of this, it is very common nowadays to use non-traditional data from different sources. The economic policy uncertainty (EPU) index is the indicator most frequently used to quantify uncertainty and is based on topic modeling of newspapers. In this paper, we propose a methodology to estimate the EPU index that incorporates a fast and efficient method for topic modeling of digital news based on semantic clustering with word embeddings, allowing us to update the index in real time, which is something that other studies have failed to manage. We show that our proposal enables us to update the index and significantly reduce the time required for new document assignation into topics.