Nonparametric, Computer Intensive Statistics: A Primer
- Conference
- Location
-
Austin, Texas
- Publication Date
-
June 14, 2009
- Start Date
-
June 14, 2009
- End Date
-
June 17, 2009
- ISSN
-
2153-5965
- Conference Session
- Tagged Division
-
Computers in Education
- Page Count
-
11
- Page Numbers
-
14.912.1 - 14.912.11
- DOI
-
10.18260/1-2--5027
- Permanent URL
-
https://peer.asee.org/5027
- Download Count
-
518
Paper Authors
Trent McDonald West Inc.
Dr. Trent McDonald is a Consulting Statistician and Senior Manager at Western EcoSystems Technology, Inc. in Laramie, Wyoming.
David Mukai University of Wyoming
Dr. David Mukai is an associate professor of civil engineering at the University of Wyoming in Laramie, Wyoming.
Abstract
NOTE: The first page of text has been automatically extracted and included below in lieu of an abstract
Non-Parametric, Computer-Intensive Statistics: A Primer
Abstract
The authors have developed a first course in statistics for engineers based on non- parametric, computer-intensive (NPCI) statistical methods. These methods do not rely on calculus or knowledge of statistical distribution theory, and as such can be taught earlier in a curriculum, are more intuitive, are less-recipe driven, and can be retained longer than traditional parametric statistics. In this paper, we provide a primer on NPCI methods. Basic NPCI concepts of bootstrapping and permutation are described. These concepts are then applied to confidence interval construction and hypothesis testing. Several examples taken from the course are worked to elucidate the methods.
Introduction
The authors have developed a new type of entry level statistics course focused on non-parametric computer-intensive (NPCI) statistics. NPCI methods do not rely on calculus because they do not depend on assumed distribution functions (thus non-parametric), instead their theory relies heavily on simple sampling concepts and their implementation utilizes computer re-sampling (thus computer- intensive). As a first course in statistics, NPCI methods are more useful for many students than traditional statistics because the basic theory posits that sampling from a sample of observed data mimics sampling from a conceptual (or real) population.
The potential benefits of a NPCI course are threefold. First, the course can be taught earlier in a curriculum than traditional statistics. Second, the methods are more intuitive and therefore stay with students longer. Finally, more sophisticated statistical procedures can be taught and used. These benefits mean that students are better equipped to solve statistical problems later in their careers. The benefits of NPCI are being investigated and results are presented elsewhere. This paper focuses on the concepts, methods, and applications of NPCI statistics.
NPCI Concepts
The theory behind many NPCI methods is not new. Many of the basic concepts have been in the statistics literature since the 1940’s. However, NPCI methods
McDonald, T., & Mukai, D. (2009, June), Nonparametric, Computer Intensive Statistics: A Primer Paper presented at 2009 Annual Conference & Exposition, Austin, Texas. 10.18260/1-2--5027
ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2009 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015