Scribd
Upload
51 views

Basic C Programming Skills Test: N: P 1 + R + R + R

This document provides instructions for four basic C programming exercises to test programming skills before starting a course. The exercises cover: 1) compiling and running a simple program, 2) using variables and control structures to compute a discounted payments formula, 3) using arrays and procedures to compute binomial coefficients recursively, and 4) adding a feature to the second program to halt if a maximum value is exceeded. Students should complete the exercises and submit the code for the first two programs before and after adding the new feature.

Uploaded by

sabtp85
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
51 views

Basic C Programming Skills Test: N: P 1 + R + R + R

This document provides instructions for four basic C programming exercises to test programming skills before starting a course. The exercises cover: 1) compiling and running a simple program, 2) using variables and control structures to compute a discounted payments formula, 3) using arrays and procedures to compute binomial coefficients recursively, and 4) adding a feature to the second program to halt if a maximum value is exceeded. Students should complete the exercises and submit the code for the first two programs before and after adding the new feature.

Uploaded by

sabtp85
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Basic C programming skills test

This is a sequence of programming exercises to be completed before you


start classes in September. You should bring the assignments with you when
you come to New York, either on a floppy or printed. This should include four
programs together with sample output. If you bring a floppy, make sure the C
code is written in four separate files in ASCII (i.e. character or .txt) format. It
is very important that you take this seriously. Not only is the project worth 5%
of your grade in the course Computing in Finance, but the course instructors
(Kishor Laud and Lee Macklin) will assume programming ability at this level
from the start of the course. We firmly believe that anyone who has been
admitted to the program will have no difficulty learning C at this level.
If you already know C programming, these should be very easy. If you
do not, you can do these as you read through a book on programming. The
exercises progress through the aspects of programming in the order they occur
in Kernighan and Ritchie.

1. Compiling and running a program


This is just to make sure you have the mechanics of editing and running a C
program. Write a simple program, based on the “Hello, world” example, that
produces the output:
(my name)
Financial Mathematics
Courant Institute
New York University

2. Variables and control structures


To understand that you know how to use variables and control structures,
write a for or while loop to compute the total discounted payments value of
n:
P = 1 + r + r2 + · · · + rn−1 . (1)
Do not use the formula for the sum of a geometric series (except possibly when
you are checking that your program gives the right answer). In your program,
compute rk using rk = r ∗ rk−1 rather than the pow function. Give the program
the values of n and r using a #define statement. The program should print the
values of n and r, for example, in the format
The discounted value of 20 payments with r = .900 is 8.7842

When this is done, add to your program a feature that halts the program
and prints an error message if P exceeds a given value PMAX, with PMAX defined
in a #define statement. Now, if I use #define PMAX 10 and take r = .95, the
program should produce something like:
P exceeded 10 with 13 payments, r = .950

1
Do this by adding a few more lines of code. You should hand in the before
(without the feature) and after (with the feature) programs. On the job, you
will probably spend more time adding features to existing programs than writing
programs from scratch.

3. Arrays and procedures


To do this, you will need to understand arrays, pointers, and the ways pro-
cedures communicate information. The normalized binomial coefficients are
n!
B(n, k) = 2−n , (2)
k!(n − k)!

for k in the range 0 ≤ k ≤ n. They can be computed using the relationship that
is the basis of Pascal’s triangle:
1
B(n, k) = (B(n − 1, k) + B(n − 1, k − 1)) . (3)
2
The formula (3) holds for k = 1, · · · , n − 1, and needs to be supplemented with

B(n, 0) = 2−n , and B(n, n) = 2−n . (4)

When n = 1, the conditions (4) define B completely. In this exercise, you will
use the recurrence relations (3) and boundary conditions (4) to compute values
of B(n, k). Do not use (2) in your program. The main ingredient is a procedure
NextB that computes all the values B(n, 0), . . ., B(n, n) (the values for a fixed
n and all integers k in the range 0 ≤ k ≤ n), given that the values B(n − 1, k)
for 0 ≤ k ≤ n − 1 have already been computed. Your procedure should have
signature
double NextB( double B[], int *firstK, int *lastK, int n);
When the procedure is called, the array entry B[k] should hold the value B(n −
1, k) for k in the range 0 ≤ k ≤ n − 1. When the procedure returns, B[k]
should have the value B(n, k) for k in the range 0 ≤ k ≤ n. To do this, the
procedure will use both (3) and (4). The procedure should return the value of
the largest binomial coefficient at level n, that is, Bmax = maxk B(n, k). For
large n, most of the binomial coefficients are very small relative to the largest
one. The largest one is for k about n/2, and they get smaller as k increases
or decreases from that value. When the procedure returns, firstK should hold
the first k value with B(n, k) > .01 ∗ Bmax and lastK should have the last k
value with B(n, k) > .01 ∗ Bmax .
Test your procedure by having it called by the following main program. It
may be interesting to see just how many large B values there are for a given n.
Feel free to experiment with larger values of NMAX.

2
double NextB( double B[], int *firstK, int *lastK, int n);

#define NMAX 100

int main() {

double B[NMAX+1];
double maxB;
int n, firstK, lastK,

B[0] = .5; /* Start by putting in the binomial */


B[1] = .5; /* coefficients for n=1, and k=0 and k=1. */
/* Why do you get a different answer using */
/* 1/2 instead of .5? */

for ( n = 2; n <= NMAX; n++ ) {


maxB = nextB( B, &firstK, &lastK, n);
printf(’n = %5d has max = %10.2f’, n, maxB);
printf(’, and large values between %5d and %5d.\n’, firstK, lastK);
}
return(0);
}

Hints and comments:

• We are using just one array, while the formula (3) uses two arrays (one
for level n and one for level n − 1). The statement

B[k] = .5*(B[k] + B[k-1]);

will produce the wrong answer. You need to save the old B(n, k − 1).
• To avoid computing 2−n for B(n, 0) and B(n, n), you can use the relation
2−n = .5 · B(n − 1, 0).
• A simple way to find firstK and lastK is to use a while loop, or a for
loop with a break statement, after the B(n, k) have been computed. That
is, loop over k at least twice.
• The normalization factor 2−n in (2) allows us to compute to much larger
values of n without overflowing the range of floating point arithmetic.
• The binomial tree method for pricing simple stock options leads to a pro-
gram that is very similar to this one.

You might also like