How Anne Hathaway Once Tricked Data Analytics to Increase Berkshire Hathaway’s Stock Value!
By Aditi Bhat
Are you pursuing a big data analytics certification course? Then you should know that the algorithms used for decoding and interpreting data can sometimes churn out unexpected results. As these algorithms are near perfect and the scope of data is huge, errors that aren’t really errors, but unexpected correlations pop up in the end result appearing seemingly illogical.
In short, data analytics can make money for you in ways that you can’t even imagine! Unfortunately, this opportunity went to the one already rolling in dough – Warren Buffet.
Fuzzy logic in data science - Emotion to algorithm
One such case was noted a few years ago by Dan Mirvish, a Huffington Post blogger. Dan saw a strange connection between the Hollywood celebrity Anne Hathaway and Warren Buffett’s Berkshire Hathaway’s stock values. Every time Anne Hathaway made it into the news, Berkshire Hathaway’s stock value went up too. Dan cross-checked the data back to 2008 to find the correlation consistent. Was Anne’s sentimental value on internet controlling Berkshire Hathaway’s stock value?
Error or opportunity?
Mirvish deduced that the algorithm used by the automated trading software was unable to differentiate certain content that was generated for Anna Hathaway and was affecting the data results for Berkshire Hathaway stock values too. Now can this accidental discovery be used to make profit?
Source: FT Alphaville
Many data scientists are attracted to such ‘strange correlations’ that algorithms throw out. As the control values for an algorithm cannot govern every aspect of data, these correlations point out to situations that data scientists and marketers can’t even think of. Serendipity strikes when these strange correlations hint towards an innovative marketing strategy.
The language of data is numbers. The correlations are not always easily spottable. Most often these strange correlations look like a bunch of numerical values acting in the same way. Correlation cannot always mean causation, but companies want to try everything out if they can see a repeating and consistent result.
Big data analytics companies are capable of creating algorithms that can read human sentiments and emotions. Combine this with social media and you can get the sentiment of an entire community towards a certain service or product. It is said that stock markets are governed by human emotions. The bullish and bearish nature of the market portrays the emotional state of the traders. Strange correlations like the one mentioned above can open up new possibilities for stock marketing that is based on the current emotional value of the target consumers.
Getting the truth out of data with critical thinking
Students undergoing big data analytics training cannot rely blindly on algorithms for every data related decision. It has to be evenly matched with a discerning human logic as to why certain results are showing strange deviations. In rare cases, these correlations can turn out to be profitable and pose as a unique algorithm strategy, to say the least.
The mantra to succeed with data is to strike a balance between logic and emotion, that is, the statistical truth complimented with human understanding.
Data analytics – The new marketing guru
Data analytics is hacking and slashing data to create meaningful, unique and profitable insights. New marketing and business strategies are seeing the light of the day armed with these insightful correlations.
Data analytics has drastically reduced the customer wait time at Disneyland and also helped Germany win the Football World Cup by helping players optimize ball possession time. It’s even able to differentiate if a toddler or an adult is using a phone, by studying just one gesture. It’s safe to say that data science is our new marketing guru in many ways.
All in all, data science courses can help professionals take challenging roles in the field and build a successful career as a data scientist. Keep working hard in this ever-burgeoning field and who knows, the data analytics tool you build someday can make you money in ways you couldn’t even imagine!