Many stockbrokers have asked the question over and over; “Can online consumer sentiment around brands be used to predict Market Movements?”
The answer is yes.
Research by Bollen, Mao and Zheng of Indiana University and the University of Manchester, used the
collective “mood” expressed by recent tweets on brands to predict whether the Dow Jones Index would decrease or increase.In their article “Twitter Mood Predicts the Stock Market”, Journal of Computational Science, Volume 2, they used sentiment analysis to classify tweets as expressing a positive or negative mood about the economy. Using Neutral networks (an analytics model), they correctly predicted the direction of change in the Dow 84% of the time. This is truly amazing because it is a widely held hypothesis that the daily market index cannot be predicted by more than 50% accuracy.
The good news is, while the research was done for the DOW, a local research was also being done to show that consumer sentiments can be used to predict Nigerian Stock Market daily movements.
Intelligent Interactive Limited, a leading digital analytics company in partnership with digital insights platform repnalytics.com, pulled data on leading brands in Nigeria (over 200,000 tweets dating as far back as 2018 was pulled) using this as a source for the research. The research was done using multiple regressive analysis (an analytics model) to predict the correlation between Consumer sentiment and Market sentiments.
A correlation was found with an accuracy above 85%, predicting market movement for top 20 brands in the Market who make up over 85% of NSE Market cap. The analytics Model gave a p.Value of 0.012 in the first data set prediction instance, and 0.047 in the second.
Data from Twitter on the following brands were used in the Research (Dangote Cement, MTN, Airtel, Gtbank, Zenith Bank, Access Bank, Bua Cement, Nestle, Stanbic IBTC, Larfarge, Nigerian Breweries, FBN Holding, Seplat, etc.). A correlation was found between brands’ positive sentiment and their daily movement, while negative, neutral, day of the week, month, year, time of day had no correlation with daily movement.
What does this mean for you as a stockbroker?
You need to keep an eye on the daily Twitter mentions and moods expressed around brands you invest in so that you can make better judgment on daily changes (this is in addition to your other intelligence or investment intelligence sources). A brand with constant positive “Mood” as expressed via Twitter, will most likely do well at the end of the Quarter or Year too.
If your firm’s name is listed below or if you are not in stockbroking but interested in the report, kindly fill this FORM to get the document via email. You will also get your username and password access to the free trial of the Twitter online sentiment tool for stockbrokers, which gives you the capacity to monitor the Daily Twitter sentiment of the top 30 brands in the market that drive 90% of the Market Cap.
List of brands with Free Website Trial access below:
No comments :
Post a Comment