The ‘Return’ and ‘Volatility’ of Sentiments: An Attempt to Quantify the Behaviour of the Markets?

Abstract

AbstractA study of ‘sentiment’ about the booming Celtic Tiger, a term associated with the economic performance of the Irish Republic during the 1990s and in the early part of this decade (c. 2000–2005), is presented by computing the frequency of usage of affect words that co-occur in the same news story with three terms: ‘Irish’ or ‘Ireland’, and economy. I present a method for quantifying changes in the frequency of the negative and positive words on a monthly basis: from these ‘returns’ we computed the so-called ‘volatility’ in the two time series – the standard deviation of the (logarithmic) ratio of current and past month’s frequency values. ‘Return’ and ‘volatility’ computations are typically carried out on the prices of financial instruments (e.g. shares and commodities) for estimating the risk associated with the instruments. We show that there is some agreement between periods of high-volatility in stock markets, or stock market indices to be precise, and the volatility of ‘bad’ news during the year. I have used a corpus of texts, 2.6 million words in total, retrieved from the Irish Times Digital Archive and published between 1995 and 2005 and used the Harvard Dictionary of Affect to compute the frequency of the so-called positive and negative affect words in the Dictionary.KeywordsSentiment analysisMarket volatilityCorpus linguisticsOntologyBehavioural finance