Polytechnic University of Valencia Congress, CARMA 2020 - 3rd International Conference on Advanced Research Methods and Analytics

Font Size: 
Mining News Data for the Measurement and Prediction of Inflation Expectations
Diana Gabrielyan, Jaan Masso, Lenno Uusküla

Last modified: 30-06-2020

Abstract


In this paper we use high frequency multidimensional textual news data and
propose an index of inflation news. We utilize the power of text mining and its
ability to convert large collections of text from unstructured to structured form
for in-depth quantitative analysis of online news data. The significant
relationship between the household’s inflation expectations and news topics is
documented and the forecasting performance of news-based indices is
evaluated for different horizons and model variations. Results suggest that with
optimal number of topics a machine learning model is able to forecast the
inflation expectations with greater accuracy than the simple autoregressive
models. Additional results from forecasting headline inflation indicate that the
overall forecasting accuracy is at a good level. Findings in this paper support
the view in the literature that the news are good indicators of inflation and are
able to capture inflation expecta-tions well.


Full Text: PDF