In 2013 and 2014, there appears to be a negative correlation between Tesla’s average monthly stock price and the monthly polarities of Elon Musk’s tweets. In 2012, 2015 and 2016, there appears to be a positive correlation between Tesla’s average monthly stock price and the monthly polarities of Elon Musk’s tweets. The coefficients of correlation for all my graphs, except for 2012 and 2016, are between .30 and .49. This means there is a moderate correlation between my variables. My coefficients of correlation for 2012 and 2016 is below 0.29, which means that there is a small correlation between my variables.
All my regression lines have R2 < 35%. This indicates that my independent variable (polarity) is not explaining much in the variation of my dependent variable (average monthly stock price). Furthermore, I am not certain if my independent variable is significant or not. The R2 values can be viewed by moving the cursor close to the regression lines in a browser. Nevertheless, I am still happy to see that there is some correlation between Tesla’s stock price and the sentiments of Elon Musk’s tweets, which means that my hypothesis is supported.
I expected to see both positive and negative correlations because Elon Musk’s tweets are erratic. Still, I could expand on this project is by doing my analysis on a bi-weekly basis instead of on a monthly basis. Another idea I have is to isolate Tesla-related tweets then conduct sentiment analysis on only those instead.