Quantifying Formality in Economic Texts: A Novel Adaptation of the Formality Formula for Economic Data Analysis

Automatic text summarization and formality analysis have been extensively studied in linguistics and natural language processing (NLP). However, their application to economic data remains underexplored. Economic texts, such as policy documents, financial reports, and news articles, often exhibit varying levels of formality that influence decision-making and communication. This study introduces an innovative adaptation of a formality formula tailored specifically for economic data. Our modified formula incorporates numerical values, domain-specific keywords, and weighted grammatical features to quantify the formality of economic texts. By applying this formula to diverse datasets, including central bank policy statements and social media discussions, we demonstrate its effectiveness in identifying formal and informal tones. Statistical validation reveals that our approach achieves significant improvements in distinguishing formal texts from informal ones, with applications in sentiment analysis, readability assessment, and document classification. The findings underscore the importance of adapting linguistic tools to specialized domains like economics, paving the way for more nuanced text analysis in financial and policy contexts.