Where regulatory announcements embody inside data, this is indicated in the announcement itself or by a observe to the announcement. As a result of the number of documents adjustments each day, whereas the number of entities stays constant, all NCI indices in our analyses are normalised by dividing them by the number documents in the corpus, m. We’ve statistically confirmed that the NCI is significantly above the level of fluctuations of the cohesiveness random null model (see Section 2 of the Supplementary Data ).\n\nWe adopt the terminology from 9 and treat our news-primarily based indicators (NCI variants and entity incidence) as indicators of the data supply in online media, whereas volumes of Google search queries are treated as indicators of knowledge demand.\n\nData supply indicators: cohesiveness index primarily based on all the news from NewStream (NCI), cohesiveness index primarily based only on filtered financial news from NewStream (NCI-financial), whole entity occurrences primarily based on the combination from all news documents and whole entity occurrences primarily based on strictly financial documents from NewStream.\n\nDeciding on financial documents also improves the correlation with other financial indices as shown in Figure 5 For more details concerning the number of financial documents and how it affects correlations with several other indices, see Section 3 of the Supplementary Data.\n\nThe indices used embody the NCI laptop using all documents, NCI-financial (calculated using selected financial documents) and its semantic parts, entity occurrences, the implied volatility of the S&P 500 (VIX), the realised historical and the day by day volatilities of the principle stock market indicators (S&P 500, NASDAQ a hundred, FTSE, DAX, Nikkei and Grasp Seng) and Google search query indicators (Business and Industrial, Bankruptcy, Financial Planning, Finance and Investing and Unemployment).