Peter Rabinovitch is the founder of PMR Data Consulting. He has extensive real world data science experience with some of the largest technology companies in the world, as well as the theoretical knowledge afforded by a PhD in Probability and Statistics. He has co-led Data For Good Ottawa, a group of volunteer data scientists who apply their expertise to solve the problems of local charities. He shares his expertise with MBA students at the University of Ottawa, meetup groups, and colleagues.
Our approach to data science (click the triangles to see details):
All real-world data sets contain anomalies.
It is the job of the data scientist to ensure that anomalies are caught, mitigated, and most importantly, that their impact is understood by stakeholders before business decisions are made.
Anomalies may be data entry errors, coding errors, or some facet of your business that you do not yet understand.
The important question is “what do the anomalies mean to your business’ bottom line?”
Using just one number to describe any real-world scenario is misleading.
Your business is too complex to be captured by just one metric. You also need to know a likely range for the metric (eg. a confidence interval), and how this metric interacts with other metrics.
Data insights must speak the language of your business, or their value may be lost.
If these insights aren’t expressed in terms that resonate with business objectives and decision making processes, their potential to drive meaningful action and create value may be diminished.
Guided by these principles, PMR Data Consulting specializes in helping businesses, large and small, extract the most important insights from their data.