Textual Analysis in R

Industry Finance
1

Dear RStudio community,

for my bachelor thesis, I am conducting a textual analysis of company-specific newspaper articles and their effect on stock returns.

As I do not have a lot of experience using R, I would like to tap this community in order to get to a starting point. What I have is two spreadsheets: One, containing all company-specific news over a 6-month period with their publication date. The other one contains negative as well as positive keywords (Loughran & McDonald, 2011).

What would be your approach, if you wanted to generate two values: 1) How many of 'hits' are there in a given news article containing a word from the positive list and 2) how many of 'hits' are there in a given news article containing a word from the negative list?

Thanks a lot in advance!

Best,
JLJ

2

The [{tidytext} package] (https://www.tidytextmining.com/) is suitable to this purpose.

3

Thanks a lot, this provided me with a good starting point!

1 Like