University of Minnesota
School of Statistics

Statistics 5021
Statistical Analysis

Course Information Sheet
January 22, 2003

Instructor Christopher Bingham 372 Ford Hall, 612-625-1024, kb@umn.edu
Office hours: Monday, Wednesday, Friday 2:30-4:00 or by appointment
Teaching Assistants Shr-Wei Chen 612-624-5569, swchen@stat.umn.edu
Office Hours Thursday 12:30-2:30 pm, 352 Ford Hall
Fan Yang 612-625-6844, yangfan@stat.umn.edu
Office Hours Tuesday 9:45 - 10:45, 352 Ford Hall
Lectures Monday, Wednesday, Friday 10:10 - 11:00 p.m., Akerman 319
Labs Section 2: Thursday 9:05 - 9:55, B53 Ford Hall (Computer Lab)
Section 3: Thursday 10:10 - 11:00, B53 Ford Hall (Computer Lab)
Section 4: Thursday 11:15 - 12:05, B53 Ford Hall (Computer Lab)
Text An Introduction to the Practice of Statistics, 4th Edition by David S. Moore and George P. McCabe, W. H. Freeman 2003
Required reading An Introduction to MacAnova, available as a course packet in the bookstore or as a PDF document downloadable from the Web )
Class web page http://www.stat.umn.edu/~kb/classes/5021

Material to be Covered: I hope to cover most of the material in Moore and McCabe through Chapter 12, starting with Chapter 1. I hope to have a more detailed list of readings for each week soon.

It is important that you read the assignments before the lectures and keep up to date. Statistics is unrelentingly cumulative; once you are behind, it may not be possible to catch up.

Web page: You are responsible for any information posted on the class web page. This will have links to assignments, sample exams and course information among other things. There may be some handouts that will be distributed only on the Web.

Email: I will be compiling an email address list from the official course registration information. If you receive your email at a different address, be sure that you have set up appropriate forwarding information. You can update this information on the web at https://www.umn.edu/dirtools.

Lectures: Except when giving an exam, I will be lecturing in each class period. I usually prepare overhead transparencies. They will not be available before the lecture I plan to post edited PDF copies of the overheads on the We within a couple of days of the lecture. There will be a link to the notes download pageb on the Class Web Page. You are responsible for what is taught in the lectures, even if it is not in the text.

Written Assignments: Most weeks there will be a written assignment to be collected and graded, about 13 in all. The overall combined homework grade will be the average of the 10 highest percentage grades. Homework will normally be due in class on Fridays (this is subject to change) and will be returned during your lab section.

Late homework will not be accepted.

On most assignments there will be listed optional problems, not to be turned in or graded. These will usually be from odd number problems in the text that have solutions on pages S-1 ­ S-28 near the end of Moore and McCabe.

In grading homework and examinations, both the correctness of the answer and organization of ideas and presentation count. It is important to show how you found an answer. Give intermediate results on numerical problems, rather than just writing down a final answer.

When you use a computer to do a problem, you should include enough information to show how you got the answers rather than just including the final answers from the computer.

Working with others: You are strongly encouraged to discuss the reading and assignments with other students. Research has shown that students learn more in statistics courses if they work with others. Caution: Don't depend too much on the help of others or you may end up overestimating your understanding of the material.

Notwithstanding the previous paragraph, the work you hand in must be your own.

Examinations (tentative schedule)

First Mid Term Monday February 24
Second Mid Term Friday April 4
Final Exam 8:00-10:00 am Tuesday, May 13

All work on examinations must be entirely your own. In case of violation, a grade of zero will be given on the examination, and the appropriate college office will be notified.

Grades: In computing the final grade, the in-class midquarter examinations, final examination, and homework will be weighted 35%, 40%, and 25%, respectively. If you are registered S/N at least a "C" is required for "S". The homework grade will be the average of the highest 10 homework percent grades.

Incompletes: According to CLA policy an "I" grade requires that "the instructor 1) has a 'reasonable expectation' that the student can successfully complete the unfinished work on his/her own no later than the end of next quarter and 2) believes that legitimate reasons exist to justify extending the deadline for course completion. ... Students are expected to make up the work on their own; if completion of the work requires that the student attend the class in substantial part a second time, assigning an I is not appropriate.".

An "I" grade will be given only for compelling reasons, and then only if a written explanation from a physician or other relevant person is provided and there is a reasonable chance you will be able to make up the course in the near future.

Computer use: From the very beginning it is required that you use a computer for doing calculations, even when you could do them on a calculator or by hand. The recommended computer program is MacAnova, a free interactive statistical analysis program. MacAnova is available for Windows 95/98/NT/ME/XP and Macintosh. A Unix/Linux version is also available.

Use of MacAnova is not required, but you are required to understand MacAnova output. I will often use MacAnova output in lecture examples and in handouts. On exams, problems may be given in which basic information is provided in the form of a transcript of a MacAnova session. Hence you ar strongly urged to bite the bullet and start using MacAnova from the start.

Reading the Introduction to MacAnova course packet and other handouts on computer use is as important as any of the other reading assignments.

If you decide to use another program, it is essential that the output or results you turn in are completely understandable by the TA, even if he is completely unfamiliar with the program you use. The basic goal is that he can tell immediately how you got the answers you did. WeI will not be able to provide help if you run into problems using another program.

The most recent MacAnova version is release 1 of Version 4.13, released on January 15, 2003. If you have any version older than release 4 of Version 4.12 (February, 2002), you should upgrade to the latest since handouts and other documentation may use features not available on earlier version.

Here is information on the various versions. You are strongly urged to get your own copy (see below for downloading information). I cannot guarantee that an up-to-date version will be available on lab computers (except for the lab in Ford B53). Any version you are likely to find on lab computers will be able to do the computations; however, some features are available only in the latest version.

MacAnova for Windows

This has up to 8 input/output windows and 8 graph windows. There are menus for switching between windows, saving files and quitting, and copying to or pasting from the Clipboard, among other capabilities.

Appendix D to the MacAnova manual has information on using MacAnova for Windows.

MacAnova for Macintosh

This comes in three flavors ­ one for Power Macintosh (PPC, G3 or G4, runs in Classic mode in Mac OS X), one for a traditional Macintosh with a 68020 or better chip with a math co-processor (will not run on Power Macs or without a co-processor) and one for a traditional Macintosh without a math co-processor (should run on any Mac). It has up to 9 command/output windows and 8 graph windows and menus similar to the Windows version.

Appendix B to the MacAnova manual has information on using MacAnova on a Macintosh.

Incorporating MacAnova output in your homework

The Windows and Macintosh versions allow you directly to copy high resolution graphs to the "clipboard" for pasting into word processor documents. The Macintosh version also allows you to save graphs in PICT or encapsulated PostScript format.

The Windows and Macintosh versions also allow you easily to use the Clipboard to import data from and export data to a spreadsheet program such as Microsoft Excel. Many users find this the easiest way to get their own data into MacAnova. See Section 5.5 in the Introduction to MacAnova.

If you have difficulty including high resolution graphs in homework, you can make low resolution graphs composed entirely of printable characters by including keyword phrase dumb:T on all plotting commands.

Printing of windows is weak point. It usually causes the Windows version to crash. For this reason, the latest Windows version has printing disabled. And Macintosh printing gives you no control of margins, headers or page numbers. On both platforms, it is much better to keep a word processor (Microsoft Word is fine) document window open while running MacAnova. You can copy text or high resolution graphs using Copy on the Edit menu and then paste them into the word processor document using Paste on the Edit menu.

All versions implement the spool() command (Spool Output to File on the Macintosh File menu). This records everything you type and everything MacAnova prints, including low resolution but not high resolution plots, in a file. Type help(spool) for detailed information or see Section 5.2 of An Introduction to MacAnova. This is another way to preserve your work for including in homework.

To get you started, you should have An Introduction to MacAnova. It is filled with examples. A good way to use it is to try to duplicate the examples yourself.

There is extensive on-line help in MacAnova. Section 4.6 in An Introduction to MacAnova explains how to use it. In addition, the help can be read on-line from http://www.stat.umn.edu/macanova/htmlhelp/help.htm

There will be handouts illustrating various aspects of the use of MacAnova.

Obtaining MacAnova 4.13 for Windows or DOS computers

Go to http://www.stat.umn.edu/macanova/downloadibm.html and click on Windows and extended memory (32 bit) DOS versions. That gets you to a page from you download installers. The minimum you need is Mv4131wx.exe, the installer for the Windows version. I recommend you also download Mv4131ht.exe, the installer for the HTML version of the help files. It's probably easiest to download these to your desk top and run them from there. You will end up with a MacAnova folder on the Program menu.

You should use the right mouse button in selecting these items so as to download them rather than display them. If you get a screen full of garbage on any of the items, try again clicking with the right mouse button.

Obtaining MacAnova 4.13 for Macintosh

Go to http://www.stat.umn.edu/macanova/downloadmac.html and then click on one of
  1. PowerPC, G3 or G4 version
  2. No coprocessor version (for pre-Power PC Macintosh)
  3. Coprocessor version (pre-Power PC with 68020 chip or better)

Then, on the new page click on the link immediately following the line starting "Every thing you need for ...". Depending on which type of Macintosh, this will be one of the following

1.MV413ppcAll.1.sea.Hqx

1.MV413nocoprocAll.1.sea.Hqx

or

1.MV413coprocAll.1.sea.Hqx

Netscape and Explorer and some other browsers will normally completely unpack this into a single folder inside of which is a folder named MacAnova4.13R1 which contains all the files you need.

Data files
Special files for use with MacAnova can be downloaded from a link on the class web page. There will be a handout on their use soon. It is not recommended that you use data files on the CD ROM which comes with the text.

Other information
You will need a hand held calculator for doing calculations during examinations. It should be able to take square roots and logarithms. A calculator which computes standard deviations and other statistical functions may be helpful if you know exactly what it computes and how it can be used, but it is not necessary. However, on homework and exams, you are expected to show how an answer was found. It is not sufficient to say, for example, "I computed the standard deviation to be 15.321."


All course materials are available in alternative formats upon request. Please contact Dana Tinsley, School of Statistics, Ford 313, Minneapolis Campus, 612-625-8046.

If you believe you should have special arrangements for exams or lectures because of a disability or impairment, you should contact Disabilities Services for assistance (612-626-1333, http://www.disserv.stu.umn.edu)


C. Bingham kb@stat.umn.edu

Updated Fri Jan 31 07:50:08 CST 2003