Impact of Elon Musk’s Tweets on Tesla’s stock price
Trading and Investing
In this research, we examine if Elon Musk’s tweets have a significant impact on Tesla’s stock price.
Because it is widely recognized that Elon Musk loves using Twitter as a platform to promote his ideas. And there is a prevailing notion that many fluctuations in Tesla’s stock price can be attributed to his fairly-say aggressive statements. As the CEO of Tesla, Elon Musk’s active engagement on Twitter and his history of making statements that impact the market are well-documented. Analyzing the impact of his tweets on Tesla’s stock price can offer valuable insights into market dynamics and the influence of prominent individuals on stock performance.
Data Resources and Summary Statistics
We need two main data parts: Elon Musk’s tweets and Tesla’s stock price.
For Elon Musk’s tweets, I used a data package available on Kaggle. Because if I want to download the tweets directly, it costs a lot from third parties. The package provides a collection of Elon Musk’s tweets spanning from January 28, 2022, to October 26, 2022, and has already done some preparation to clean the text data. It contains a total of 2,668 raw data entries.
Table 1. Raw Tweets Data
To ensure relevance to Tesla, I implemented a filter to extract tweets specifically related to Tesla and got 422 data entries. Then, I generated a word cloud to visually represent the most prominent words in the collected data. In addition, conducted sentiment analysis using VADER. We can see from the sentiment score chart that there are not many trends to find.
Fig. 1. Word Cloud with Musk’s tweets
Fig. 2. Sentiment Score Over Time Fig. 3. Tesla Stock Price 2022/02 -2022/11
For Tesla’s stock price, I directly extracted the adjusted closing price from Yahoo Finance and ensured that the start date and end date for collecting stock price data matched the timeframe of the tweets data..
OLS and Result interpretation : Impact of Elon Musk’s Tweets on Tesla’s stock price
To examine the relationship between sentiment scores of Elon Musk’s tweets and the adjusted closing price of Tesla stock, here I applied OLS regression. Because it allows for examining the relationship between two variables and determining the strength and direction of the association. In this case, the independent variable was the sentiment score of Elon Musk’s tweets, which represents the overall sentiment expressed in his tweets. The dependent variable was the adjusted closing price of Tesla stock.
OLS Results:
• R-squared value is 0.034, indicating that only about 3.4% of the variation in the stock price can be explained by changes in the sentiment score. This suggests that the sentiment score alone may not be a strong predictor of Tesla’s stock performance.
• The coefficient for the sentiment score is -25.9942. It indicates that, on average, a one-unit increase in the sentiment score is associated with a decrease of approximately $26 in the adjusted closing price of Tesla stock.
• The p-value for the sentiment score is 0.051, suggesting that the relationship between the sentiment score and the stock price may not be statistically significant at the conventional significance level.
Table 2. OLS Results
Conclusion and Reflection
In conclusion, the research findings indicate a weak association between the sentiment score of Elon Musk’s tweets and the adjusted closing price of Tesla stock from January 2022 to October 2022. This outcome contrasts with my initial expectations but seems reasonable upon further consideration. Other studies, however, suggest a stronger correlation between Elon Musk’s tweets and Tesla’s stock price. Some researchers propose that Musk’s tweets have a more significant impact on trading volume, subsequently affecting the stock price.
Reflecting on my research, I recognize several limitations that may have contributed to the nonsignificant results:
• The quality and time span of tweets: although I tried other musk’s tweets sources covering past five years, the results didn’t change much. But I do have doubts on the data packages’ quality.
• The sentiment scoring system employed, however, not designed explicitly for Elon Musk or tweets. Musk’s unique language style and the informal nature of tweets make it harder to give accurate judgments.
• The temporal relationship between tweets and stock prices: to examine the potential influence of tweets on stock prices, it is imperative to enhance the clarity and refine the methodology for delineating the temporal dynamics of tweets and stock prices. Specifically, it is crucial to establish a sequential ordering where tweets precede any potential impact on stock prices. However, due to the underwhelming outcomes obtained thus far. As a result, further analysis has become withheld at this stage. I will improve on the above points if I can find more solid tweet sources.
In conclusion, conducting this research and applying the techniques learned from your course is a great pleasure. Lastly, I really enjoyed this research! Lastly, lastly, enjoy!
The New York Stock Exchange | NYSE
