what's new with analytics in academia?
DESCRIPTION
Building the Analyst of the Future. Presentation by Jeff Camm, Director of the center for Business Analytics at the University of Cincinnati.TRANSCRIPT
UP NEXT… 3:00pm
What’s New with Analytics in Academia?
Building the Analyst of the Future
DR. JEFF CAMM
Follow the action on Twitter using #AtE2014
Interest in Analytics
2
What's New with Analytics in Academia?
Building the Analyst of the Future
Jeffrey D. Camm
Director, Center for Business Analytics University of Cincinnati
Lindner College of Business Department of Operations, Business Analytics & Information Systems
4
5
Why now?
l Big Data l Better Software l Better/cheaper computing
We create as much information in two days now as we did from the dawn of man through 2003.
6
l Social Media l GE Aviation l dunnhumby l IRI l Healthcare
Big Data
7
Competing on Analytics
Some companies have developed a corporate-wide analytical mindset and are now competing based on analytics.
8
Our working definition: Analytics is the scientific process of transforming data into insights for making better decisions. This includes descriptive, predictive and prescriptive models.
What is Analytics?
9
What does it mean to be scientific?
The Scientific Method – Ask a Question – Do Background
Research – Construct a Hypothesis – Test Your Hypothesis
by Doing an Experiment
– Analyze Your Data and Draw a Conclusion
– Communicate Your Results
The Engineering Design Process
– Define the Problem – Do Background
Research – Specify Requirements – Brainstorm Solutions – Choose the Best
Solution – Do Development Work – Build a Prototype – Test and Redesign
Source:
Source: 10
Source:
Source:
11
l Descriptive – what happened? l data queries, reports, descriptive statistics,
data visualization
l Predictive – what will happen? l linear regression, time series analysis, data
mining, simulation
l Prescriptive – what should we do? l optimization, simulation/optimization,
decision analysis
Categorization
12
Descriptive Analytics
13
Predictive Analytics
Cincinnati Zoo: l # Donors = 0.0213*(Zip Code Population) – 26.941
– For every increase of 100 people in a zip code, we expect about 2 more donors
– Adjusted R2 = 0.3847
l # Donors = 0.0196*(Zip Code Population) + 0.0026*(Avg Home Price in Zip Code) – 372.15 – For every $1000 increase in average home price in a zip
code, we expect about 2.6 more donors – Adjusted R2 = 0.4857
14
Prescriptive Analytics
l $1B + NPV l $250M
savings per year
North American Product Supply Study
15
Analytics Maturity
Source: SASSAS 16
What will be the life cycle of this movement?
17
McKinsey Report
By 2018, the U.S. could face a shortage of 190,000 data scientists and another 1.5 million managers and analysts who know how to use big data to make effective decisions.
18
l Evangelists (me J)
l Enablers (Analytics Graduates)
l Consumers (Management)
Gartner defines 3 Analytics Personas:
19
How has academia responded to the
demand for analytics?
20
l Enablers (Masters Programs in Analytics)
l Consumers (MBA core courses, electives in analytics, MBA tracks)
21
New Programs:
22
Data Informed’s Map of University Programs in Big Data Analytics
23
24
25
Source: NC State 26
UC MS-Business Analytics
27
Our MS Business Analytics Program
28
Curriculum for Enablers
Based on Klimberg, Business Intelligence, INFORMS 2011 (Hinrichs, SEDSI, 2012)
29
Prerequisites:
UC MS Bus Analytics l Multivariate Calc. l Linear Algebra l Programming l Business Core
NC State MS Analytics
l Statistical Methods l Regression l Statistical Computing &
Data Management
30
UC Electives (Basic Business Knowledge)
31
Core Courses:
UC MS Bus Analytics
l Probability Modeling l Statistical Methods l Data Management l Statistical Computing l Statistical Modeling l Optimization Modeling l Simulation Modeling l Optimization Methods
NC State MS Analytics
l Analytics Tools and Techniques
l Analytics Foundations l Analytics Methods &
Applications I l Analytics Practicum I l Analytics Methods &
Applications II l Analytics Practicum II
32
NC State:
33
UC Electives (10 credit hours)
34
l Individual Project l Case Studies in Analytics l Internships
UC Capstone
35
l Some are more focused: – Northwestern: MS Predictive Analytics – UCONN: MS Business Analytics and
Project Mgt. – Wash U. St. Louis: MS Customer
Analytics
Other Programs
36
l Starting Salaries: – $65k to $135K – Virtually 100% placement
l Positions – Analyst – Data scientist – Application Area Specific
Payoff
37
l Software vs Methodology
l Consulting vs Analyst
Possible Pitfalls
38
l What’s the difference? – Business Knowledge? – Hard Coding? – Statistics – Optimization and Simulation? – Traditional vs Machine Learning
Analytics vs Data Science
39
Columbia: MS Data Science
30 credit hours
40
l Methods for organizing data, e.g. hashing, trees, queues, lists, priority queues. Streaming algorithms for computing statistics on the data. Sorting and searching. Basic graph models and algorithms for searching, shortest paths, and matching. Dynamic programming. Linear and convex programming. Floating point arithmetic, stability of numerical algorithms, Eigenvalues, singular values, PCA, gradient descent, stochastic gradient descent, and block coordinate descent. Conjugate gradient, Newton and quasi-Newton methods. Large scale applications from signal processing, collaborative filtering, recommendations systems, etc.
Columbia: MS Data Science
Algorithms for Data Science
41
Analytics vs Data Science
Source: Jerry Smith, datascientistinsights.com
42
l Analytics 1.0 - - the era of “business intelligence.
l Analytics 2.0 - - big data analytics (with small math)
l Analytics 3.0 - - the intersection of the two, with every company joining the data economy
What does the future hold?
Source: Davenport
43
§ Mixture of data types § More analytics than in the 2.0 big data world § Everything faster—technology, methods § Analytics baked into processes and decisions § Chief Analytics Officers emerge § Analytics become prescriptive § Data science gets mixed in § Many data integration options
Analytics 3.0
Source: Davenport
44
Thanks!
Questions?
45