The last article focused on what regression was and how the results can be interpreted. It mentioned that there were a number of assumptions required in order for the model to be valid. The assumptions are necessary because they relate to the reasons why a regression line works well as a prediction. The assumptions are based on the residuals, which are the difference between the predicted value of the dependent variable in the regression and the actual y value in the regression. Continue reading Understanding Regression – Part 2
Understanding Regression – Part 1
Decision makers are always looking for ways to understand the effects of their actions. Managers generally assume that if they find a correlation between two items it means they understand the relationship between two variables; however, as was stated in a previous blog article, Beware of Correlations, correlations may not tell the whole story, and, furthermore, they can only tell the story between two variables. Regression allows us to understand more involved relationships between variables and an outcome variable.
Conjoint Analysis: Finding Customer Preferences
Marketers are always looking for ways to identify which attributes of products their customers really like. Customers tend to choose products, but subconsciously they rate their preferences based on the attributes of those products. These attributes, also called stimuli, affect the customers decision making behavior.Conjoint analysis is a technique that allows the researcher to ascertain the value customers place on different attributes of a product or service, without specifically asking about the attributes themselves.
Continue reading Conjoint Analysis: Finding Customer Preferences
Challenges for Data Science Organizations
Its easy for large companies to spend large budget dollars on multi million dollar data initiatives, thereby taking risks that could yield significant rewards. Small to medium size companies dont always have the luxury of a large budget to build their analytics group or solution, and in many cases dont have the time, or patience, to wait for a solution to yield results. Large companies, while having the luxury of a budget, suffer from inertia, sometimes preventing progress.
Should you trust the numbers?
Recently NBC reported that the that the offensive coordinator of the Tampa Bay Buccaneers, Dirk Koetter said, “I dont need a bunch of numbers, … I trust my eyes”. Many individuals in organizations feel the same way. Sometimes they don’t trust the numbers. Why should they, as Mark Twain said, there are “Lies, damn lies, and statistics.” Any information obtained, will have a use, whether you use it properly is a completely different question. Sun Tzu said, “Know thy enemy and know thyself and you need not fear the result of a hundred battles”.
Ultimately, the key to success is studying and learning about yourself and your opponents. Your eyes can deceive you because, while your eyes can see what is happening your brain interprets the images received, and bias can set in. Countless times in history, events are clearly seen, however, people and leaders fail to act because of cognitive paralysis, cognitive bias, or cognitive distortion.
“Know thy enemy and know thyself and you need not fear the result of a hundred battles”.
Data and analytics provides an unbiased approach, so long as the analyst conducts the analysis in an unbiased manner. You cannot discount the success of professional sport teams that use data to their advantage. The Oakland Athletics, shown in the movie “Moneyball”, are probably the most famous at using analytics and data to change their way of thinking and thus removing bias. Many critics argue that while they have been successful in having a lower payroll, it has not translated into World Series Championships. However, the Boston Red Sox, under General Manager Theo Epstein, utilized the data to obtain a roster that would ultimately defeat the New York Yankees in what might be considered one of the most memorable comebacks in baseball history, and culminate their cinderella run defeating the St Louis Cardinals in 2004, ending their 86 year drought. Many other teams in baseball such as the Arizona Diamondbacks, New York Mets, and St Louis Cardinals all have very deep statistical departments, as do many other teams in different sports.
What shouldn’t be a surprise is that the analytics by themselves cannot win championships. Everything is a process and a give-and-take between the intuition and the mathematics (See the Yin Yang of Analytics). There are gut feels based on environment and changes in the landscape that a leader must take into account; however a good leader will never underestimate an opponent and would never want to overlook information. Sun Tzu again sais that “it is only the enlightened ruler and the wise general who will use the highest intelligence of the army for purposes of spying and thereby they achieve great results.” In this case, the spies are the virtual spies, or the data that yields the intelligence unknown previous to the leader, and in many cases unknown to the opponent.
Consider always that information is useful, and the analysis and analytical process is a fundamental part of sports teams, and every organization. Without having a quality set of analytics, you will be at a disadvantage and thus increase your chances of losing; with good analytics you set the stage for victory and thus “not fear the result of a hundred battles”.