Can we analyse sentiments within financial terminology?

5092

Sentiment analysis

“Cause you know sometimes words have two meanings”, sing Led Zeppelin in their masterpiece ‘Stairway to Heaven’. And this must have been what Bill McDonald had in mind when, together with Tim Loughran, he wrote his Master Dictionary.

“A growing body of finance and accounting research uses textual analysis to examine the tone and sentiment of corporate 10-Ks, newspaper articles, press releases, and investor message boards”, states the opening of their essay ‘When is a liability not a liability? Textual analysis, Dictionaries and 10-Ks’.

McDonald and Loughran chose to focus their attention on 10-Ks ˗ a comprehensive summary report of a company’s performance that must be submitted annually to the Securities and Exchange Commission in the US.

Among the thousands of words comprising the Dictionary, there are some that typically have negative, positive or other implications in a financial sense, and that are analysed according to a discipline known as sentiment analysis. Out of these words they created six lists that they renamed Fin-Neg (a list of negative words), Fin-Pos (positive words), Fin-Unc (words that express uncertainty), Fin-Lit (words reflecting a propensity for legal contest, i.e. litigiousness), MW-Strong and MW-Weak (strong and weak modal words, respectively).

The idea that lies behind the creation of these lists is that there are “significant relations between stock price reactions to the sentiment of information releases as measured by word classifications”, says McDonald.

So what is sentiment analysis? Known also as opinion mining, it is defined in Wikipedia as referring “to the use of natural language processing, text analysis and computational linguistics to identify and extract subjective information in source materials”.

Since it first became popular in 2001, countless papers and essays have been written on this topic, and projects such as McDonald’s dictionary have started. Its growth coincides with the ever-growing popularity of social networks, blogs and forums, which comprise a high volume of digital data now available for analysis.

One of these projects is SSIX, financed by the EU, which has among its main objectives the assigning of sentiment to financial texts which “are extracted from several sources of social networks, news sites and blogs, in several languages”.

For further information on McDonald’s Master Dictionary and its Word Lists check this page.

 

Written by Maria Bregolato
Terminology trainee at TermCoord