Pros And Cons Of Econometric Modeling
Episode 03
Episode 03 Transcript
Welcome again to the Media Mix Minute. I’m John Colias of the Advanced Analytics Group at Decision Analyst. In an earlier episode, I-we explained that media mix modeling uses historical data to measure the ROI of media activities, such as, digital marketing, or TV advertising. I also explained that econometric modeling is the primary traditional approach used in media mix modeling. Today we will discuss the key strengths and weaknesses of econometric modeling when used for media mix modeling.
First, econometric modeling—it reduces the bias in measurement. That’s its first advantage. Second, it correctly or accurately isolates out the impact of the media (the impact of media on sales) from the impact of all of the other factors that influence sales. So both of these key advantages are very important, and it’s while it’s... beyond the scope of what we would discuss in terms of all the detail about this, suffice it to say that econometricians and statisticians have, for decades, developed methods that accomplish these benefits. And the approach (the use of econometric modeling and media mix) it utilizes these benefits.
Every method has its advantages and has its disadvantages as well, as does econometric modeling. And the key disadvantage to econometric modeling, is that if two types of media are highly correlated in the historical record (that is, they move together our over time), then it reduces the ability of the econometric modeling to isolate out and separate the impact of each media type on sales independently becomes reduced. In addition, the ability of econometric modeling to project to levels of media activity beyond the historical record is limited. Of course, with these limitations the obvious solution to this would be that we could inject variability into the historical data in order to mitigate these disadvantages of econometric modeling. But that would be the subject of a future Media Mix Modeling episode. Thank you!
Presenter
John Colias, Ph.D.
Senior VP Research & Development
As a leader with both university teaching and business consulting experience, John focuses on predictive modeling, prescriptive analytics, and artificial intelligence. As Senior Vice President, Research & Development, at Decision Analyst, John combines academic and business interests to help analytics professionals by offering cutting-edge analytic solutions tempered by business realism. He holds a doctorate in economics from The University of Texas at Austin, with specializations in econometrics and mathematical modeling methods.
