Robert A. Stine

Robert A. Stine
  • Professor Emeritus of Statistics

Contact Information

Research Interests: credit scoring, model selection, pattern recognition and classification, statistical computing and graphics, time series analysis and forecasting

Overview

Education

PhD, Princeton University, 1982
MA, Princeton University, 1979
BS, University of South Carolina, 1977

Recent Consulting

Fraud detection in loan applications; validating models in use for consumer credit default.

Career and Recent Professional Awards

Miller-Sherrerd MBA Core Teaching Award, 2003, 2006, 2007, 2010
David W. Hauck Award for Outstanding Teaching, 2001
Excellence in Teaching Award (Undergraduate Division), 2001, 2004

Academic Positions Held

Wharton: 1979-present (Research Associate, Analysis Center for Evaluation of Energy Modeling and Statistics, 1979-83, Director of Computing Analysis Center, 1979-83).

Previous appointments: Princeton University; University of Michigan; University of South Carolina. Visiting appointment: University of Michigan

Other Positions

Summer Intern, Office of Energy Information Administration, U.S. Department of Energy, 1978

For more information, go to My Personal Page

Continue Reading

Research

Teaching

Current Courses (Spring 2024)

  • STAT4220 - Predictive Analytics

    This course follows from the introductory regression classes, STAT 1020, STAT 1120, and STAT 4310 for undergraduates and STAT 6130 for MBAs. It extends the ideas from regression modeling, focusing on the core business task of predictive analytics as applied to realistic business related data sets. In particular it introduces automated model selection tools, such as stepwise regression and various current model selection criteria such as AIC and BIC. It delves into classification methodologies such as logistic regression. It also introduces classification and regression trees (CART) and the popular predictive methodologies known as random forest and boosted trees. By the end of the course the student will be familiar with and have applied these concepts and will be ready to use them in a work setting. The methodologies are implemented in a variety of software packages. Applications in JMP emphasize concepts and key modeling decisions. This course may be taken concurrently with the prerequisite with instructor permission.

    STAT4220001 ( Syllabus )

  • STAT7220 - Predictive Analytics

    This course follows from the introductory regression classes, STAT 1020, STAT 1120, and STAT 4310 for undergraduates and STAT 6130 for MBAs. It extends the ideas from regression modeling, focusing on the core business task of predictive analytics as applied to realistic business related data sets. In particular it introduces automated model selection tools, such as stepwise regression and various current model selection criteria such as AIC and BIC. It delves into classification methodologies such as logistic regression. It also introduces classification and regression trees (CART) and the popular predictive methodologies known as random forest and boosted trees. By the end of the course the student will be familiar with and have applied these concepts and will be ready to use them in a work setting. The methodologies are implemented in a variety of software packages. Applications in JMP emphasize concepts and key modeling decisions. This course is formerly STAT 6220.

    STAT7220001 ( Syllabus )

All Courses

  • STAT1020 - Intro Business Stat

    Continuation of STAT 1010 or STAT 1018. A thorough treatment of multiple regression, model selection, analysis of variance, linear logistic regression; introduction to time series. Business applications. This course may be taken concurrently with the prerequisite with instructor permission.

  • STAT3990 - Independent Study

    Written permission of instructor and the department course coordinator required to enroll in this course.

  • STAT4050 - Stat Computing with R

    The goal of this course is to introduce students to the R programming language and related eco-system. This course will provide a skill-set that is in demand in both the research and business environments. In addition, R is a platform that is used and required in other advanced classes taught at Wharton, so that this class will prepare students for these higher level classes and electives.

  • STAT4220 - Predictive Analytics

    This course follows from the introductory regression classes, STAT 1020, STAT 1120, and STAT 4310 for undergraduates and STAT 6130 for MBAs. It extends the ideas from regression modeling, focusing on the core business task of predictive analytics as applied to realistic business related data sets. In particular it introduces automated model selection tools, such as stepwise regression and various current model selection criteria such as AIC and BIC. It delves into classification methodologies such as logistic regression. It also introduces classification and regression trees (CART) and the popular predictive methodologies known as random forest and boosted trees. By the end of the course the student will be familiar with and have applied these concepts and will be ready to use them in a work setting. The methodologies are implemented in a variety of software packages. Applications in JMP emphasize concepts and key modeling decisions. This course may be taken concurrently with the prerequisite with instructor permission.

  • STAT7050 - Stat Computing with R

    The goal of this course is to introduce students to the R programming language and related eco-system. This course will provide a skill-set that is in demand in both the research and business environments. In addition, R is a platform that is used and required in other advanced classes taught at Wharton, so that this class will prepare students for these higher level classes and electives.

  • STAT7220 - Predictive Analytics

    This course follows from the introductory regression classes, STAT 1020, STAT 1120, and STAT 4310 for undergraduates and STAT 6130 for MBAs. It extends the ideas from regression modeling, focusing on the core business task of predictive analytics as applied to realistic business related data sets. In particular it introduces automated model selection tools, such as stepwise regression and various current model selection criteria such as AIC and BIC. It delves into classification methodologies such as logistic regression. It also introduces classification and regression trees (CART) and the popular predictive methodologies known as random forest and boosted trees. By the end of the course the student will be familiar with and have applied these concepts and will be ready to use them in a work setting. The methodologies are implemented in a variety of software packages. Applications in JMP emphasize concepts and key modeling decisions. This course is formerly STAT 6220.

  • STAT7240 - Text Analytics

    This course introduces modern text analytics, and the tools of natural language processing. Text and language are powerful repositories of knowledge and information, but the semi-structured nature of language makes deriving insights from text challenging. Modern analytic techniques introduced in this course make it significantly easier even for non-specialists to use text and language data to drive deep insights. The course will use several examples from real world applications in different industries such as ecommerce, healthcare and finance to illustrate these techniques. Students should be familiar with regression models at the level of Stat 6130 or Stat 1020, and the Python language at the level of Stat 4770 or Stat 7770. Familiarity with the Jupyter notebook development environment is presumed, as well as common Python packages such as pandas, NLTK and SpaCy. Those with more knowledge of Statistics, such as from Stat 7220/4220, or computing skills will benefit. The predominant software used in the course is Jupyter notebooks that use a Python interpreter. Familiarity with basic probability models is helpful but not presumed.

  • STAT8990 - Independent Study

    Written permission of instructor, the department MBA advisor and course coordinator required to enroll.

  • STAT9950 - Dissertation

  • STAT9990 - Independent Study

    Written permission of instructor and the department course coordinator required to enroll.

Awards and Honors

  • Wharton MBA Excellence in Teaching Award, 2012
  • Helen Kardon Moss Anvil Award for MBA Teaching, 2011
  • Wharton MBA Excellence in Teaching Award, 2009
  • Wharton MBA Excellence in Teaching Award, 2006-2007
  • Excellence in Teaching Award (Undergraduate Division), 2004
  • Miller-Sherrerd Wharton MBA Core Teaching Award, 2003
  • David W. Hauck Award for Outstanding Teaching, 2001
  • Excellence in Teaching Award (Undergraduate Division), 2001
  • Miller-Sherrerd Wharton MBA Core Teaching Award, 1999
  • Wharton MBA Core Curriculum Teaching Award, 1996
  • Excellence in Teaching Award (Undergraduate Division), 1995
  • Miller-Sherrerd Wharton MBA Core Teaching Award, 1994-1995

Activity

Latest Research

Sivan Aldor-Noiman, Lawrence D. Brown, Emily Fox, Robert A. Stine (2017), Spatio-temporal Low Count Processes with Application to Violent Crime Events, Statistica Sinica, 26, pp. 1587-1610.
All Research

In the News

More Savings, Less Plastic: Consumer Credit after the Crisis

The harsh economic downturn that has chastened credit-happy consumers, along with increased scrutiny by regulators, will force card issuers to rethink their business models as the economy begins to recover, according to Wharton faculty and credit industry analysts.Read More

Knowledge at Wharton - 7/8/2009
All News

Wharton Stories

Sivan Aldor Noiman writing on a blackboardHow This PhD Student Discovered a Dynamic Research Community at Wharton

Sivan Aldor-Noiman, GR’12, came to Wharton Doctoral Programs from a small statistics program in Israel, and found herself in a dynamic center of intellectual life….

Wharton Stories - 11/13/2014
All Stories