Spotkanie nr 12 / Meetup #12

This is a past event

57 people went


[EN] Everyone is invited! If someone doesn't understand Polish and wishes to join the meeting let us know and the presentations will be given in English.

Presentation: "How to apply R to real-life business problems", 1 hour + Q&A

This time McKinsey Knowledge Center invites you to join us to learn how R and advanced analytics can make great impact in real business cases. Come and learn more about:

* How McKinsey Analysts use R to drive modern advanced modeling and analytics projects
* Mathematical methods to find the best solutions for business problems
* Career opportunities with McKinsey in Wroclaw

After our session, we invite you to stay for a casual Q&A discussion over some snacks and soft drinks with our team members.

Summary of two business cases:

1. Visualizing Multidimensional Data with R.
The Organizational Health Index is a solution that enables companies to measure and identify the organizational phenomena that constitute their identity. Working with clients for more than 10 years, we have gathered answers from more than 3 mln respondents from all over the world. As Data Scientists, we help teams to extract knowledge and insights from our data.
Recently, we were asked to visualize how companies are distributed across 37 practices we measure. Requirements made cost function highly non-linear: results have to be explainable for everyone, and they have to be pleasant to look at which significantly complicated the task. During the session we will show you dimensionality reduction techniques which we used for solving the problem.

2. Using R to risk-adjust performance of healthcare providers.
Healthcare industry is on its way to link payments for services to its quality. To be able to do that payors must build complex analytical systems that will track a range of quality metrics, assess and compare performance of providers and finally tie that to payments. One of the key challenges on that road is to make sure that the comparisons are fair, i.e. they are adjusted for differences in factors which correlate with performance: a hospital treating the elderly will have a higher mortality rate than a hospital treating younger patients, but to be able to make an apples-to-apples comparison the performance should be adjusted by age.
We helped our client to build a performance measurement system that takes into account differing population risk factors. This involved defining quality metrics, statistical modelling of performance, and implementing analytical layer on top of their data warehouse. We used R to test various models and to create a proof-of-concept of methodology, which was later implemented into the system.

For more information about McKinsey in Wrocław please visit: