Preface

This compendium constitutes the main reading for the course Log 708 - Applied Statistics at Molde University College. This course is in its current version a mandatory course for a number of master programs in logistics and business administration at the college. The wide variety in backgrounds of the students in the course, as well as the particular selection of topics for the course has made it difficult to find appropriate literature among the available commercial titles. In particular, the tight connection between theory (statistics) and practice (using R software) that we strive to achieve in the course made it tempting to put together a targeted text like this.

The document is typeset with the Rbookdown typesetting system using Rstudio, R and MikTeX on a Windows PC. I thank those who have developed and maintained these beautiful and free software tools. Thanks to Till Tantau for making the TiKZ and Beamer packages for LaTeX.

Some colleagues contributed: Katerina Shaton provided valuable feedback on preliminary versions of the text, and Yauhen Maisiuk did a great job chasing out numerous misprints and errors in the previous version of the text. Thanks to both of them. Deodat E. Mwesiumo revitalized chapter 1 by adding several examples. In addition he has contributed substantially to the exercise material for the course (available through other sources).

The course title has two key words which will serve to lay out the direction of the course.

Statistics The core topics of the course are within the field of statistics. Statistics as a subject is often held by students to possess a rather toxic combination of mindboggling boredom and immense difficulty. Here we hope to deal with the material in a way that minimizes any of those two aspects. Can we do that without compromising the nature of the subject? Statistics undeniably involves a bit of mathematics and certain crucial concepts that need to be understood in order to gain working skills within the subject. This text aims at providing students with the adequate skills and tools for using statistics in a reliable way.

Applied The students in the course will come with a variety of different academic backgrounds. With the assumption that most of the students of this course are not great lovers of mathematical formalism and technical detail, we will try to always keep a close connection to relevant applications of our methods. This guides the choice of examples as well as the selection of exercise material.

There is no way of doing applied statistics without access to suitable software. In Log 708 we have chosen to use R/Rstudio as the main software tool. The goal is that students during the course will gain familiarity with this software so as to make it a natural choice for data analysis throughout the MSc program. We try to make practical R examples for most theoretical aspects introduced, so as to stay in line with the applied nature of the course.

Prerequisites

The ideal background for this course is for the students to have completed a course in basic statistics at bachelor’s level. Traditionally such a course would include general probability and an introduction to the normal probability distribution. A basic introduction to parameter estimation, with the concept of confidence intervals would also normally be part of such a course. If you also have some previous knowledge about testing of statistical hypotheses, it certainly would not hurt. That’s the ideal situation. Presumably there will be students in the course that lack all or parts of an ideal background. For those students, an extra effort should be put in as early as possible to try to close the gap. There will be recommended reading material that students with little statistical background can use as a supplement to get started. Also, the lectures will start with a quick review of required background material. Previous experience shows that a motivated student who is willing to throw in an extra effort early in the semester will have no problem of completing the course in good shape. This has been seen even for students with very little previous exposition to statistics.

All students in this course will have completed a three or four year study program equivalent to a European bachelor’s degree. Being admitted to a MSc study program means you are likely among the more successful students in your previous program. You will now be an experienced student knowing how you learn best. In the MSc program you will be even more responsible for your own learning, with an ability to identify where and when you need to put in extra efforts. An additional skill is the ability or willingness to communicate with teachers and fellow students about challenges in a subject. If something is really difficult to understand or accomplish, you should try to discuss with someone, and ideally resolve the problems. Otherwise, “holes” in your knowledge of a subject may make further learning difficult. All of these “general student skills” can also be considered an ideal prerequisite for starting in a master study program.

The only absolute prerequisite is a willingness to work with a positive attitude and to do your best in the course, based on your background and starting level.

List of topics

The following topics are central to this text

  1. Basic descriptive statistics.
  2. Basic inferential statistics with hypothesis testing.
  3. Regression analysis.
  4. Using R within all the above-mentioned topics.

Supplementary literature

There is a large number of text books covering much of our material in greater detail. Almost any book you can find with words like “business statistics”, “introductory statistics” or “elementary statistics” in the title will cover the topics 1 and 2 and much of what we will do on regression analysis. One recommended text for those who want a supplement is (Paul Newbold 2005) (See references at the end of the compendium), where most of chapters 1 - 10 are relevant to topics 1 and 2, chapters 11 - 13 are relevant to topic 3.

A good alternative is (William L. Carlson 1997) which is similar to (Paul Newbold 2005) in level and scope. Further textbooks that can supplement on the basics of the course are (Richard D. De Veaux 2010), (Levine et al. 2010), (Robert Stine 2011) and (McClave, Benson, and Sincich 2010). As of writing, all of these are available in the college library.

References

Levine, D. M., M. L. Berenson, T. C. Krehbiel, and D. F. Stephan. 2010. Statistics for Managers (Using Microsoft Excel). Any. Pearson. http://books.google.no/books?id=fKkuRAAACAAJ.
McClave, J. T., P. G. Benson, and T. Sincich. 2010. Statistics for Business and Economics. 7.–11. ed. Pearson. http://books.google.no/books?id=2-TtAAAAMAAJ.
Paul Newbold, Betty Thorne, William L. Carlson. 2005. Statistics for Business and Economics. 6., 7. or 8. Pearson.
Richard D. De Veaux, David E. Bock, Paul F. Velleman. 2010. Intro Stats. Any. Pearson.
Robert Stine, Dean Foster. 2011. Statistics for Business (Decision Making and Analysis). Pearson.
William L. Carlson, Betty Thorne. 1997. Applied Statistical Methods - for Business, Economics and the Social Sciences. Prentice Hall.