Social media posts and stock returns: The Trump factor
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Purpose This paper explores the effects of US President Donald Trump's Twitter messages (tweets) on the stock prices of media and non-media companies. Design/methodology/approach The authors’ empirical analysis considers all Twitter messages posted by Donald Trump from May 26, 2016 (the date he passed the threshold of 1,237 delegates required to guarantee his presidential nomination) to August 30, 2018. The authors accessed President Trump's tweets through http://www.trumptwitterarchive.com , which provides links to all Twitter messages the President has ever posted. Of the 6,983 presidential tweets during our sample period, the authors select 513 messages that mention companies that are publicly traded in the United States for this study. The selected messages are then classified as having a positive, neutral or negative sentiment. The authors employ a series of univariate and multivariate tests as well as Heckman two-step regressions and partial least squares regressions to examine the effect of the President's tweets on the stock prices of the firms he tweets about. Findings For media firms, the authors find that positive tweets have a pronounced positive stock price impact, whereas negative and neutral tweets have little or no effect. For non-media firms, the authors observe the opposite: negative tweets tend to be associated with significant stock price declines, whereas neutral and positive tweets incur weakly positive stock price reactions. To a large extent, these stock price declines reverse on the following day. The authors further find that the President's reiteration of information that is already known by the market incurs an additional stock price reaction. The President's attitude towards the news appears to play a major role in this context. Originality/value The authors contribute to the literature by offering various new insights regarding the effect social media has on the stock markets. In addition, this paper expands the emerging strand of literature that explores how President Trump affects the stock prices of firms he tweets about. This paper differs from prior studies in this area by considering a broader range of tweets, by controlling for potential selection biases, by differentiating between Trump's tweets about media and non-media firms and by exploring the impact of “old” vs “new” news based on whether the President repeats information that is already known to the market. If social media posts by single influential people are found to affect markets, they may create trading opportunities for investors and financial managers and risk arbitrage opportunities for arbitrageurs. In the political science field, the findings of this research provide valuable insights into how politicians can employ social media platforms to affect the public, and the differential influence of nominees and politicians in office. Finally, our study gives corporations that wish to back a certain campaign or a candidate in an election a better idea of the possible risks and benefits of their actions, considering that candidates or politicians could post negative messages on social media platforms targeting companies that backed their opponents.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it