Saturday, January 23, 2016

This week the AIM students will learn about ‘Big Data Analytics’ and how to use it to conduct fundamental stock analysis


Can the monitoring of real-time Twitter feeds be used to identify trends that lead future stock price movements?


The seniors in the AIM program have begun the spring semester with a module on behavior finance (see previous blog).  They are learning that emotions can profoundly affect individual behavior and decision-making. This is different than neo-classical finance theory which suggests that people are rational and always make the best informed decisions.

Can these two competing theories co-exist? Do individuals in a society experience emotions or mood states that affect their collective decision making? Can we tap into the public sentiment and determine in advance emerging trends?
Image result for twitter
Fundamental stock analysts want to know as early as possible what’s "hot" and what isn’t - what products and services are trending positively? What's hot: Apple Watch or Pebble Time; Beats or Skullcandy; Lego Friends or Zoomer Dino; Chipolte or McDonalds; Yoga pants or Levis; Facebook or Instagram; etc.)? Which products are gaining and or losing market share?
Dr. David Krause, AIM director
Fundamental analysts would love to know in advance what is the current public mood regarding a company's goods or services before this information is reported by the firm. They would like greater predictive accuracy of a firm's future success – in other words, can we use non-traditional, non-financial information from social media to help predict increasing or weakening consumer demand for specific products or services?
Is this a way to get a jump on future firm success or weakness? Dr. David Krause, AIM program director, believes so. "We've finally started using Sabermetrics in making better decisions in baseball, why shouldn't we be using advanced metrics in assisting the equity analyst perform fundamental analysis?"

"Data is growing by an amazing rate with the increased use of text messaging, emails, blogs, open source surveys, Facebook, Twitter, etc. The challenge is to access, array and analyze this also overwhelming amount of information - and to use it real-time to make educated investment decisions," according to Krause.

"This week we are going to introduce the AIM students to the use of big data sources and the tools that can be employed in an attempt to understand consumer trends, Krause stated.

Terence Thong-Hwee Ow
Dr. Terence Ow
Google Trends, Google Analytics and IBM Watson Trend are examples of free ‘big data analytic’ websites that can be used to monitor macro trends. "We should also dive into various the various social media databases (i.e. Twitter, Facebook, LinkedIn, etc.) to extract raw text feeds and analyze it for micro trending." According to Dr. Krause, "Detection of the Chipolte e. coli outbreak in advance of published reports could have resulted in a substantial trading opportunity." 


Alex Isken
Alex Isken
Dr. Terence Ow, Associate Professor of Management at Marquette University and Alex Isken (a member of the AIM Class of 2015) will be presenting the results of Alex’s Twitter Data and Stock Sentiment Analysis of Tesla to the AIM students next week. A paper that has been submitted recently to an undergraduate research competition. 

Previous research has indicated that it is possible to investigate real-time measurements of collective mood states derived from large-scale Twitter feeds to predict the sentiment or mood about a firm’s products or services. 

Alex Isken, under Dr. Ow's tutelage, specifically examined whether it was possible to use real-time social media feeds to identify trends that are correlated and lead future stock price movements. This will be new material that has not been presented to the students in the AIM program to this point. 

`These students will learn how to extract and array data from Twitter and how to analyze the information to create a company-specific sentiment index. This 'Twitter sentiment' will be analyzed over time to determine whether it can be used to enhance traditional fundamental equity analysis.

"Determining whether a rising or falling Twitter sentiment trend can serve as a predictor of a company’s future relative stock price is going to be an enhancement to the existing curriculum," according to Dr. Krause. It will be an interesting week.