Skip to content
All library documents

Testing Whether a Large Negative Stock Return Has a Lasting Impact

Article Quant Q&A · Author: alexbougias

Summary

The document asks how to use time-series analysis to determine whether an unusually large negative stock return represents a lasting price impact or fades over a longer horizon. It frames the shock as a decline several standard deviations beyond the mean and suggests a physical disaster affecting a company in a vulnerable sector as a possible event to study.

The question motivates an event-impact analysis, but gives no proposed statistical test, model, comparison group, or empirical evidence. It also leaves important choices unresolved, including how to define abnormal returns, estimate the expected-return baseline, and distinguish a persistent shock from later market or firm-specific movements. The suggested horizon is illustrative rather than a tested result, so the document is best treated as a research question rather than a complete method.

Key ideas

  • The research question is whether an extreme negative stock shock persists or dissipates over time.
  • A disaster affecting a sensitive sector is offered as a possible event setting.
  • A time-series approach is requested, but no specific test or model is given.
  • Defining abnormal returns and controlling for other price drivers remain unresolved.

Tags

Full text
# Negative abnormal stock return and permanent impact


# Negative abnormal stock return and permanent impact












Assume we have a day where stock price falls many standard deviations of the mean (e.g >3) . How could we test, in terms of time-series, if this negative shock is permanent or deminishes in the long run(e.g 1 year).

One example, would be testing the impact of a physical disaster on a stock, component of a sensitive to these phenomenon sector.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.