Scribd
Upload
74 views4 pages

Mathematical Association of America Mathematics Magazine

This document summarizes a paper that presents an algebraic and geometric derivation of Cramer's Rule for solving systems of linear equations using cross products. The algebraic approach writes the system of equations in matrix form and solves using Cramer's Rule. The geometric approach views the coefficients as defining a plane and uses cross products of vectors in the plane to derive a normal vector, relating it to Cramer's Rule. The argument is generalized to higher dimensions by replacing cross products with determinants.

Uploaded by

88915334
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
74 views4 pages

Mathematical Association of America Mathematics Magazine

This document summarizes a paper that presents an algebraic and geometric derivation of Cramer's Rule for solving systems of linear equations using cross products. The algebraic approach writes the system of equations in matrix form and solves using Cramer's Rule. The geometric approach views the coefficients as defining a plane and uses cross products of vectors in the plane to derive a normal vector, relating it to Cramer's Rule. The argument is generalized to higher dimensions by replacing cross products with determinants.

Uploaded by

88915334
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Using "Cross Products" to Derive Cramer's Rule

Author(s): Donald R. Barr


Source: Mathematics Magazine, Vol. 38, No. 2 (Mar., 1965), pp. 64-66
Published by: Mathematical Association of America
Stable URL: http://www.jstor.org/stable/2688422
Accessed: 10-04-2017 01:09 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

Mathematical Association of America is collaborating with JSTOR to digitize, preserve and extend access
to Mathematics Magazine

This content downloaded from 170.140.142.252 on Mon, 10 Apr 2017 01:09:54 UTC
All use subject to http://about.jstor.org/terms
This content downloaded from 170.140.142.252 on Mon, 10 Apr 2017 01:09:54 UTC
All use subject to http://about.jstor.org/terms
1965] USING "CROSS PRODUCTS 1 TO DERIVE CRAMER'S RULE 65

A. Algebraic Approach. Since the three points must satisfy equation (1), we
have a system of three linear equations in the unknowns a, ,B, and y. This system
can be written

(2) AcT = 1 T
where A is a matrix whose i-th row is composed of the coordinates of the i-th
given point, c=-(a, 3, y),1 (1, 1, 1), and the "T" indicates "transpose." Of
course, this system can be solved using Cramer's Rule.

G. Geometric Approach. A vector normal to the plane can be obtained by


taking the cross product (also called the vector product or outer product) of
two nonzero, nonparallel vectors u and v in the plane. Since the coefficient vec-
tor c is also normal to the plane, it follows that uXv=Xc, where X is a real num-
ber. Knowing Xc, we can easily obtain the coefficients in equation (1).
The observation that (A) and (G) both lead to the solution of equation (2)
suggests the possibility of proving Cramer's Rule using cross products. (Note
that any system of three nonhomogeneous linear equations in three unknowns
may be put in the form of equation (2).)

Case for E3. Suppose (x1, yi, z1), (X2, Y2, Z2), and (X3, Y3, Z3) are three noncol
points on a plane whose equation is given by (1). We seek the solution to the
system (2). The vectors U=(X2-X1)i+ (Y2-y1)j + (Z2-z1)k and v = (x3-xl)i
+(y3-y1)j+(z3-z1)k are in the plane, where i, j, and k are the usual orthogonal
unit basis vectors for E3. Thus, using the usual determinant expansion for cross
products, we have that

i j k

(3) u X v-=Xai + X3j + Xyk= x2-xl Y2-Yl Z2-Zl


X3-Xl Y3-Yl Z3-Zl

is a vector normal to the plane.


Multiplying equation (1) through by X and equating multiples of i, j, k in
equation (3), we have

(4) [(y2-yl)(Z3-Z1) -(y3-yl)(Z2-Zl)]X+ [(X3-Xl)(Z2-Zl) -(X2-X1)(Z3-Zl)]y

+ [(x2-xl)(y3-yl)-(X3-Xl)(y2-yl)]Z=X,

where X is obtained by letting x = x1, y = yi, and z = z1 in equation (4). It is ea


verified by expanding the respective determinants that equation (4) may be
written as

(5) IAlI x + I A21 y + ? A3 z = I A I,


where I AijI is the same as I A j except that the j-th column has been re
by a column of ones.
Comparing equations (1) and (5) we have a- IA11/IAI if |A| 10 with
similar expressions for ,B and y. This is, of course, Cramer's Rule for the system
(2). A geometrical condition on the plane which insures that system (2) is non-

This content downloaded from 170.140.142.252 on Mon, 10 Apr 2017 01:09:54 UTC
All use subject to http://about.jstor.org/terms
66 MATHEMATICS MAGAZINE [Mar.-Apr.

homogeneous and has nonzero coefficient determinant is that it does not contain
the origin; i.e., it has nonzero intercepts.

Special case for E2 and generalizations. The argument used above is easily
specialized to E2 and generalized to higher dimensional spaces, provided that we
merely replace the cross product arising in E3 by a determinant of appropriate
order. These cases are also interesting in that they point out a "cross product"
technique of finding a vector in En normal to a given n -1 flat. One can easily
prove this normality using the fact that determinants with one or more repeated
rows are zero. It is also easy to show that the geometrical condition on the flat
insuring the existence of the nonhomogeneous system and nonzero coefficient
determinant remains the same under change in dimensionality of the space
considered.

TANNERY'S THEOREM

R. P. BOAS, JR., Northwestern University

The title of this note is presumably as unfamiliar to most American mathe-


maticians as it was to me when I encountered it recently. Tannery's theorem
for series ([1], p. 123; [2], p. 136) deals with limits such as lim 0 (1 +x/n)n and
([4], p. 467)

lim + + +

and even with a derivation of the power series for the sine and cosine without
using Taylor's formula. It says that if fk(n)->Lk for each k, as n->oo, and if
fk(n) ?Mk with EMk convergent then
co

fl (n) + f2 (n) + * * fp (n) -- E- Lk;

provided that p-* oo as n->oo. Bromwich remarks ([2], p. 136), " . . . the test
for the theorem is substantially the same as the M-test due to Weierstrass ....
The proof, too, is almost the same." It is a good test of a student's grasp of uni-
form convergence to ask him to verify that the analogy here is extremely close:
the theorem is a special case of the M-test. (Cf. [3], p. 122.)
There are similar theorems for infinite products and for improper integrals.

References

1. P. L. Bhatnagar and C. N. Srinivasiengar, The theory of infinite series, National Publish.


ing House, Delhi, 1964.
2. T. J. I'A. Bromwich, An introduction to the theory of infinite series, 2d ed., Macmillan,
London, 1926.
3. E. W. Hobson, The theory of functions of a real variable and the theory of Fourier's series,
2d ed., vol. 2, Cambridge, 1926.
4. C. A. Stewart, Advanced calculus, 3d ed., Methuen, London, 1951.

This content downloaded from 170.140.142.252 on Mon, 10 Apr 2017 01:09:54 UTC
All use subject to http://about.jstor.org/terms

You might also like