For years and years, PMs have been "little data" people. We plot data points, compute averages, look at 1-sigma, 2-sigma (and sometimes, 6-sigma) limits, ponder the Central Limit Theorem, and wonder about the law of large numbers
Fair enough
What then is "big data" if not simply more numbers, more data points?
Andrew Gelman -- eminent authority in the statistical analysis field -- has this pithy answer:
“Big Data” is more than a slogan; it is our modern world in which we learn by combining information from diverse sources of varying quality."
And, so what is it we do when the data quality varies?
- Get more, to see if there is a discernible and useful pattern
- Use Bayesian techniques to refine hypothesis based on observations and feedback
- Throw out the obviously bad stuff, though try not to throw it out just because it's an inconvenient counter-point
Not familiar with Bayes? Search this blog site; you'll find a lot of stuff here
Read in the library at Square Peg Consulting about these books I've written
Buy them at any online book retailer!
Read my contribution to the Flashblog
No comments:
Post a Comment