The effect of news on stock prices: AAPL

For public companies, the stock price is one facet of their reputation that is very easily tracked. Following on from a number of conversations I’ve had recently, I have become more and more interested in the effect off news (and gossip and buzz) on share prices.

At some point, I hope to be able to publish a case study on the effect of blog buzz on share price - I think this will be likely to come about only when we have happened to have been monitoring a public company’s reputation through a crisis or release of some great news. It is hard to reconstruct these kinds of effects after the fact.

A starting point, though is to look for correlations between a company’s share price movements and online news volumes. In order to find an interesting case study for this, I looked through the recent histories of a few shares and settled on the period in January this year when Apple (symbol: AAPL) released the iPhone. This is interesting because shortly after the release of information that exerted an upward pressure on the share price, Cisco sued over the name (which is identical to the trademarked name of one of their own products - the companies have since agreed to kiss and make up).

Google finance shows a good summary of the performance on NASDAQ of AAPL through the year and most of the data used in this analysis is sourced there.

News volumes in January 2007

The chart below shows online news volumes for major stories (along with associated sentiment) through January 2007 for Apple. It has been normalised so that the day of the greatest discussion shows as ‘1’.

aapl-news

AAPL stock price movements in January 2007

In order to do my best to remove effects from the wider market, I took the closing prices for AAPL stock through Januray 2007 and rebased them to a starting point of 100. I did the same for the value of the NASDAQ index over the same period. Subtracting the NASDAQ movements from the AAPL movements left those movements that can be genuinely attributed to Apple. The data you see plotted below is the movement of the stock price (so a positive number is an increase the price). In order to be able to compare it side-by-side with news volumes, this was also normalised so that the greatest movement in any given day shows as ‘1’.

aapl-stock

News and stock price movements side by side

Plotting these two sources of data side by side is interesting:

aapl-analysis

As my stats professors would like to remind me (and as the freakonomics guys tell us) , correlation is not evidence of causality. I am not trying to say that online news sources caused the share price movements we see (the underlying story and offline media will obviously play a major role). It is interesting (and as it should be, in an efficient market) that practically all the major movements in the share price are accompanied with equivalent news.

I am looking forward to being able to do this analysis in real time with blog buzz volumes as there is some evidence that looking at intra-day movements, blog buzz can actually come before the share price movement.

Will Critchlow

Will Critchlow

Will founded Distilled with Duncan in 2005. Since then, he has consulted with some of the world’s largest organisations and most famous websites, spoken at most major industry events and regularly appeared in local and national press. Will is part...   read more

Get blog posts via email

7 Comments

  1. IanD

    Interesting post..key is being able to predict "something" is happening and ideally doing so as early as possible..
    Would help to know buzz volume and buzz quality (positive or negative)..

    reply >
  2. Hi Ian,

    Thanks for dropping by and commenting.

    I think you are right about the steps: first try to tell that something is happening and then track volume and quality.

    I hope we can make some more progress automating this and coming up with some really useful tools! Stay in touch for future developments!

    reply >
  3. Rahul

    I am new to the field of behavioral finance, and am doing some research on effect of news on stock prices. could you please tell me how did you measure news volume in your study ? thanks

    reply >
  4. Hi Rahul,

    It's a while since I did this analysis! If I remember correctly, I collated numbers of stories from sources such as Google and Yahoo News, de-duplicated them and counted them by day.

    Drop me an email (will@distilled.co.uk) if you want to chat about it further.

    Thanks
    Will

    reply >
  5. Just a thought....

    I think I may have a stock that would be a good canidate for testing purposes.

    I am a holder of stock in lon:CRH cement roadstone holdings.
    The shares have been in the family for years.
    I intend holding onto them for years to come through many
    ups and downs.
    They are a boring company that perform well over time in pretty consistent fashion regarding annual earnings.

    The share price has been really badly affected lately by news on
    oil, the dollar and US sub prime mortgage fears.
    Also a contributing factor to their decline is worries about Irish property prices. Ireland being where the company is based, though the company is geographically very diverse.

    News on the company of late has been pretty scarce.
    Alot of the downward price movements could be attributed
    to non-company related news.
    In the coming months the comapany will complete a huge multi billion dollar deal with Cemex. This news is already in the public arena.

    However I expect news levels to greatly increase upon completion of the deal. It would be interesting to see what affect that will have on the share price especially if there is an absence of new developments in oil, dollar etc.

    If interested I could give you some detailed figures that might flesh this idea out better.

    If anyone reading this would like to buy stock in CRH feel free :-)

    reply >
  6. I think you'd need a bigger sample size to be honest than one company if you wanted any valid findings out of it!

    reply >
  7. Mc

    I wonder how you classfy the positive & negative news?

    reply >

Leave a Reply

Your email address will not be published. Required fields are marked *

*
*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>