Date: Tue, 17 Dec 2002 18:44:28 -0500 (EST)
From: Adrian German <dgerman@cs.indiana.edu>
To: A113 Fall 2002 Distribution List <dgerman@indiana.edu>
Subject: A113 Final Grades


A113 Final Grades have been posted. Evaluation points will
be added tomorrow at noon. Please check the grades and let
me know if you see any problems. If we need to make changes
we need to make them tomorrow, before I turn the grades in to
the Office of the Registrar. So please let me know.

I want to thank you all for your dedication and hard work and
to wish you a happy break and a happy new year 2003.

... Adrian

Here's the message of yesterday:

Date: Sun, 15 Dec 2002 15:24:02 -0500 (EST)
From: Adrian German <dgerman@cs.indiana.edu>
To: A113 Fall 2002 Distribution List <dgerman@indiana.edu>
Subject: A113 Update: Grades, Exams, Evals 


Dear A113 Friends,

The semester is coming to an end. All of the grades that we have (all
of them) have now been posted and we tried to calculate as much of the
final grades as we could. A total of 33 students now have a letter grade
(most of them have taken the final exam early on Thu). The final exam is
on Tue Dec 17 12:30-2:30pm TH A201 (as also listed in INSITE). Those that
have taken the final early and are happy with their grade need not come
on Tue. Those who do come have a chance to improve their grades if they
do better than last Thu when they took the early final exam (which was
meant as a hands-on review).

The gradebook still contains some N/A's and some later's (an indication
that some grades are still being worked on). Please make sure you contact
me or any one of the TAs to clear your account of any late assignments or
exams before Tue Dec 17 at the end of the day. After this date every grade
that appears as N/A or later in the gradebook becomes a 0 (zero). I plan
to post the final grades on Wed Dec 18 and to turn them to the Registrar's
on Thu Dec 19. Please make sure you contact me for any questions you may
have early next week, to see what we can do to help, if appropriate.

Don't forget the bonus project. It's due at exam time, and it's worth one
full assignment. The text of the bonus project is listed at this URL:

http://www.cs.indiana.edu/classes/a113-dger/fall2002/notes/Eleven.html

There will be one more way to collect an extra point, which is by turning
in an evaluation for this course with the CSCI Department. The CSCI Dept.
is collecting course and instructor evaluations for all classes and A113 is
one of them. Included are instructions on how you can submit your evaluation
for A113. You must submit your evaluation before Wed Dec 18 at noon (the day
after the final exam, so I can include the bonus point. It doesn't matter
what you write in your evaluation: as long as you turn it in you get an extra
bonus point at your final grade.) I will receive a summary of usernames for
those that have submitted an evaluation on Dec 18 at 4pm so I can use the
list to add a bonus point. Please let me know if you have any questions. Here
are the instructions, as promised:

  1. Go to the following website:  http://www.indiana.edu/~best/

  2. Under "QuizSite", choose STUDENTS

  3. From the drop-down box, choose/instructor to be evaluated
                           (example: CSCI-A113-dgerman)

  4. Click PROCEED

  5. Check "DO A CURRENT ACTIVITY" (Should say "Evaluation")

  6. Click PROCEED
     Evaluation for requested instructor should be displayed

  7. Answer Questions

  8. To submit completed evaluation, click PROCEED at bottom of page
     Then wait.

     If student has multiple CS courses or course instructors, they will
     be enabled at that time to click SELECT ANOTHER COURSE button

     Please Note - Students are allowed a max 3 attempts to complete each
     evaluation.  (They will be enabled to open each individual evaluation
     3 times, whether or not they make any additions or changes.)

Please help the department (and us) improve this class by describing
what you liked and what you didn't like in the class. We thank you for
your participation and for your interest in A113. Should you have any
questions whatsoever please let us know.

Sincerely,
Adrian German, for CSCI A113 Fall 2002

Thirty-three final grades have been calculated and posted.

They are not final until after the Final Exam (Tue 12:30-2:30pm TH A201).

All N/A's and laters will turn into 0's at the end of the day (11:59:59pm) that day.

Please clear your account (late homework grades) with your TA by then.

Let me know if you have any questions or concerns.

All exams have been posted.

If your final letter grade appears as N/A it may be because:

1. some homework grade is still being calculated for you, or
2. you have not taken the final yet (which is on Tue Dec 17 12:30 TH A201)

Only about 22 grades can be computed as of this time.

Check the current situation here (A113-1285).

Remember you can retake the final and/or submit the bonus project for extra.

1. 9:30am Practical Two
2. 11:15am Practical Two

3. 4:00pm Practical Two

Exam is (of course) open book and open notes, but individual work.

(The final is still offered on Dec 17 as scheduled between 12:30 and 2:30 in TH A201. If you take it twice the highest of your two grades will be kept, so you can also take today's exam as a final exam review. Tomorrow there is a Practical exam in lab, the last one).

Here are the optional makeups we will be administering today:

1. 9:30am Practical One Makeup
   9:30am Practical Two Exam

2. 11:15am Practical One Makeup
   11:15am Practical Two Exam

3. 4:00pm Practical One Makeup
   4:00pm Practical Two Exam

Wednesday we have a Practical One make-up. Here's what to practice.

1. Lab tomorrow is to be used for review and exercises.
   (Homework Four will also be collected tomorrow in lab).
2. Tue next week we have a Midterm Makeup (notes posted)
3. A Bonus Project is included in Tue (Dec 10) notes
4. (It is worth 5%, same as a homework assignment, and is due on Dec 17)
5. Last homework assignment (posted) is due on Wed in lab
6. Wed next week you can take the Practical Make-Up (see notes)
7. (You can also take the Second Practical on Wed if you want to)
8. Thu (Dec 12) next week you can take the Final Exam
9. Dec 17 you can retake the Final (if not happy with your Dec 12 grade)

10. I will post a few more notes on the Final Exam later tonight.
11. Today in class we have a review. Let me know if you need help.

Rest of notes to be posted Wednesday night, with grades.

1. today will be the last time new material will be presented.
2. on Thu we will have a final review
3. next Tue there will be a Midterm Makeup in class
4. next Thu one can take the final exam in advance (in class)
5. Homework Four is due on Fri in lab
6. Homework Five will be due on Tue in class
7. labs tomorrow and on Fri aim to help with the last assignments
8. there will be a Practical Makeup in lab next Wednesday
9. the second Practical Exam is next Fri (last lab) in lab
10. the final is on Dec 17 but if you do well next Thu you don't need to come
11. there will also be a bonus take home project that one can turn in Dec 17

The material for this week is Chapter 6 and the beginning of Chapter 7 in Kirkup.

Date: Mon, 25 Nov 2002 22:14:05 -0500 (EST)
From: Adrian German <dgerman@cs.indiana.edu>
To: A113 Fall2002 Distribution List <dgerman@indiana.edu>
Subject: A113 Update Thanksgiving Week 


Midterm and practical grades have been posted yesterday with a small
set of practical grades to be posted shortly. Many people did well, and
some did not do that well. For those that did not do very well there will
be a chance to make up individually, in person, by individual appointment
before Dec 10. I will post the make up midterm and practical on Tue 11/26
on the web site. Both will be optional, everybody can re-take them, but
you must do this before Dec 10.

This week we start a new topic: correlation and regression. The material
for the rest of the semester will come from Chapters 6 and 7 of Les Kirkup.

If you have any questions or concerns please let me know.

... Adrian

More details to be posted soon.

1. Correlation, and
2. Regression

The other six quizzes are now in review mode (disclosing the correct answers).

Lecture notes for Dec 3 sketched.

Date: Mon, 25 Nov 2002 22:14:05 -0500 (EST)
From: Adrian German <dgerman@cs.indiana.edu>
To: A113 Fall2002 Distribution List <dgerman@indiana.edu>
Subject: A113 Update Thanksgiving Week 

Midterm and practical grades have been posted yesterday with a small
set of practical grades to be posted shortly. Many people did well, and
some did not do that well. For those that did not do very well there will
be a chance to make up individually, in person, by individual appointment
before Dec 10. I will post the make up midterm and practical on Tue 11/26
on the web site. Both will be optional, everybody can re-take them, but
you must do this before Dec 10.

This week we start a new topic: correlation and regression. The material
for the rest of the semester will come from Chapters 6 and 7 of Les Kirkup.

If you have any questions or concerns please let me know.

... Adrian

Date: Wed, 20 Nov 2002 09:31:02 -0500 (EST)
From: Adrian German <dgerman@cs.indiana.edu>
To: A113 Distribution List Fall 2002 <dgerman@cs.indiana.edu>
Cc: Fulya Erdinc <ferdinc@cs.indiana.edu>,
     Mieko Furuhashi <mfuruhas@cs.indiana.edu>, akoswara@indiana.edu,
     jasdande@indiana.edu
Subject: A113 Labs Today


Here are some questions that you may be asked in labs today
as a review for both exams this week:

a) plot the normal distribution with an average of 56.7 and
   a standard deviation of 2.3 (Hint: use NORMDIST, FREQUENCY,
   and divide the standard deviation units in 10 intervals each).

b) given this normal distribution, and a score of 59 what
   percentage of the population scored below? above? between
   54.4 and 56.7? below 54.4? (Hint: the mean and the standard
   deviation are really your ruler and compass when exploring
   a normal distribution). So this problem can be solved with
   just paper and pencil, and you don't even need that.

c) using the same normal distribution as above what is the
   score for which exactly 75% of the population has scored
   below? (Hint: use NORMINV(75, mean, stdev, 1) but why?)

A few answers to some of the posted problems:

d) given that the sum of the faces on two dice is 6, the
   probability that one face is a 4 is 2/5 (that is, 0.4).

e) in the game of 6 dice one should get $0.83255 on average for
   every $1.00 played so the game is unfair. That is to say that
   if you were to play 1000 times ($1000.00) you'd win $830.255
   so you'd lose about $169.75 (again, so you can check results)

f) for the two coin game the payoff is 7/8, that is you lose $0.125
   (on average) for every $1.00 you pay (to participate in one round).
   So playing 1,000 times ($1000.00) you'd be winning $875 roughly.

For d), e), and f) you should be able to set up the experiments
already, and I believe you need the answers to verify that they do
indeed match your findings.

As reminder the exams are open-notes and open-book. The midterm covers
mostly measures of central tendency (mean, mode, median) and dispersion
(range, variance, average deviation, standard deviation), histograms,
frequencies, relative frequencies, distributions, and probabilities as
limits of relative frequencies. The normal distribution is a special type
of distribution, defined in terms of a mean and a standard deviation of
the sequence (or set) of values. Standard (z) scores are one way of
comparing scores in distinct (different) normal distributions.

The practical wants you to develop an experiment and measure the
probability of some event. You will be asked to turn in one screen
shot of your work and you will be able to write on its back anything
that you thought would add to the screen shot in describing your
experiment and its results.

If you have any questions or concerns please let me know.

... Adrian

1. some solutions and hints
2. (not many, but essential)

to help you prepare for the exams.

Also, congratulations to Yukiko Tsujihata who correctly pointed out that

1. if we use an odd number of coins in Homework Three
2. the results of the experiment become significantly more accurate.

This is a rather technical issue, but can we see why that is so?

Homework Four and lecture notes for next week also posted.

The practical is on Friday (4pm).

What to study for the midterm?

Answer:

1. Solve all 6 (six) quizzes in QuizSite.
2. Study Chapter 2 and part of 3 from Kirkup.
3. Review web notes and be clear on the following topics: mean, mode, median, standard deviation, frequency, relative frequency, probability as the relative frequency of occurrence, distributions, normal distribution, z-scores.

Both exams are open book and open notes but individual work.

The practical will be like any of the homework assignments and lab assignments.

You might get a problem like this:

Set up an experiment in Excel to find out the probability of getting three heads when you throw four coins.

Here are screen shots that show how the golden coin experiment was done:

1. the start of the experiment
2. the end of the experiment
3. the mechanism (set of formulas)

Notice that the last picture is really big so we can see.

Also notice that for this problem to be really effective one needs to provide two fairly short, and somewhat persuasive solutions, that give different answers.

So here they are:

1. Solution One: the probability must be 1/2. The reason is: if the first coin is gold, then you are in either drawer one or drawer two. So there are two possible cases. In one of the two possible cases you get a gold coin and in the other a silver coin. So 1 in 2 is the chance to get a second gold coin if the first one is gold.

2. Solution Two: the probability must be 1/3. The reason is: you basically pick two coins and would like the coins to be both gold. You have three drawers, so the total number of cases is 3. In only one case you get two gold coins, so you have 1 in 3 chances of getting a second gold coin when the first one is gold.

This way one of them must be incorrect (perhaps both?) and a carefully chosen experiment can tell which is incorrect and give a hint as to what the right answer might be.

Remember:

1. the purpose of computing is insight, not numbers.
2. the only way we can deal with infinity is by thinking about it.

1. Screen One (the Central Limit Theorem checked)
2. Screen Two (basically Homework Three report)
3. Screen Three (drawing a normal distribution)

I hope to provide more screenshots for the other experiments, later.

1. FREQUENCY
2. IF
3. COUNTIF
4. Excel Help
5. lab experiments
6. counting with pivot tables (Tutorial 5)

(Any counting that we are doing is done in the context of our experiments).

We are getting ready to take the exams next week.

Review notes with sample exam exercises are forthcoming.

Use the help notes in Excel for FREQUENCY to set up a test of your understanding.

That is, set this up:

Suppose a worksheet lists scores for a test. The scores are 79, 85, 78, 85, 83, 81, 95, 88, and 97 and are entered into A1:A9, respectively. The data_array would contain a column of these test scores. The bins_array would be another column of intervals by which the test scores are grouped. In this example, bins_array would be C4:C6 and would contain the values 70, 79, 89. When FREQUENCY is entered as an array, the number of scores corresponding to the letter grade ranges 0-70, 71-79, 80-89, and 90-100, are counted. This example assumes all test scores are integers. The following formula is entered as an array formula after you select four vertical cells adjacent to your data. The fourth number, 2, is the count of the two values (95 and 97) that are greater than the highest interval (89). So

FREQUENCY(A1:A9,C4:C6) equals {0;2;5;2}

Reading assignment for Les Kirkup Chapter 3 is from pp. 85-121.

(Soon we are going to summarize what you need to review for next week's exams.)

Date: Mon, 11 Nov 2002 10:08:49 -0500 (EST)
From: Adrian German <dgerman@cs.indiana.edu>
To: A113 Fall 2002 Distribution List 
Subject: A113 Update


Today is Monday November 11. Here are the A113 news:

1. Wednesday in lab you can turn in Homework Two (for
the first time if you didn't turn it in on time, or a
revised version of it if you think you've had trouble
with it the first time around). No penalty is involved.

2. Homework Three is now due Friday, in lab. We'll use
most of tomorrow to clarify it further and to actually
run an adequate simulation. Then, and on Thursday too,
we'll be doing a few more simulations and probability
estimates.

3. Notes on the web will be updated Monday night for
the whole week. The Midterm is in class on Nov 21, and
is a written, and open-book, open-notes exam. We'll have
a Midterm Exam Review on Nov. 19, the lecture before the
exam. The Practical exam is in lab, on Friday Nov 22.

We'll post more details during the week. If you have any
questions or need any help please let us know.

... Adrian

Date: Fri, 8 Nov 2002 19:46:42 -0500 (EST)
From: Adrian German <dgerman@cs.indiana.edu>
To: Undisclosed recipients:  ;
Subject: A113 Update


With Homework Three we start becoming more sensitive to what exactly
the goal is in an experiment. In this case the goal is really simple
as shown in class on Thursday: can an experiment produce a distribution
that resembles a theoretical distribution closely? Well, yes, but under
what circumstances?

Remember the example discussed in class. All experiments must start
small. Then they need become as big as they need be, but not bigger. So,
in this case, can you (or should you) throw 1000 coins? The answer is: no.
The question then is: why? Well, what are you trying to achieve? That is,
can you obtain the same results with only ten coins? Twenty? Forty? 100?

I'd through 20 coins 20 times first. Plot. Estimate the fluctuations by
pressing F9 a few times. Then I'd throw 20 coins 1000 times, or 4000 times,
and only then more (like 15,000). Plot the distribution every time and turn
in the printed copy of the graph (only). Be careful not to print thousands
of lines of spreadsheet though!

The result of your report is needed, not the spreadsheet.

Increasing the number of coins increases the number of outcomes. Then the
number of times we need to run the experiment also increases so then the
problem becomes unmanageable, and for 1000 coins (hard as it is to do that
in Excel) perhaps about 10,000,000 trials would do justice to the experiment.

Getting back to the purpose of the assignment, we need to produce the normal
distribution with as little headache as possible. So the lecture on Tuesday
will show you how two dice (and a bit of knowledge, or external thinking)
would be enough. But more on that Tuesday.

Meanwhile please let me know if you have any questions or if you need any
help. I will be able to answer e-mail over the weekend, so let me hear from
you if you have any trouble working this homework out. Thanks a lot for your
interest, once again, and hope your weekend will be a great one.

... Adrian

(For a practical exam, possibly).

How can you approach such a problem. Labs this week try to help.

1. Lecture Notes for this week, Lab Notes posted.
2. Homework assignments Two and Three posted (What's Due?).
3. Lab Notes this week also help with the new assignments.
4. Two new quizzes have been posted in QuizSite.
5. Lecture Notes Five sketched (the midterm exam is coming).

Please log in to verify you can get in.

If you can't please let Adrian know (at dgerman@indiana.edu).

Lab Notes One posted.

Lecture Notes One posted.

Homework Assignment One posted, will be discussed in class.