Below I give a short summary of the contents and the most important files
The variables for the panel regressions are stored here.
These include the control variables from Compustat.
Raw data that are either directly from the source or used to give extra insights.
data_nyse_screener.xlsx
contains the (cleaned) company names from the NASDAQ Screener.
data_lemma_internet.txt
is the frequency list, note that you to multiply the 'frequency' with 181,376 to get the real total amount of occurences.
data_lexicon.xlsx
is the lexicon, the list that contains many English words.
This file contains the architecture of the neural network.
z_predictions_100x.xlsx
is the file that contains the fluency predictions for each company name.
Contains the 3 panel regressions in Eviews.
Contains Jupyter Notebook files written in python.
Contains the figures that are used in this project.
Bibtex style references to other studies.
This file contains the thesis written in LaTeX.
If you were curious why... π
-
data_
Raw data obtained from scources -
data_filtered_
Processed raw data -
data_reg_
Regression data for testing the effect of fluency on the characteristics -
z_
Input and output data that are from the deep learning model -
py_
Jupyter Notebook files with python code