MathGroup Archive 1992

[Date Index] [Thread Index] [Author Index]

Search the Archive

Problem with MatrixPower, v. 2.1

  • To: mathgroup at yoda.physics.unc.edu
  • Subject: Problem with MatrixPower, v. 2.1
  • From: "Katherine (Williams) Derbyshire" <kewms at kew.com>
  • Date: Wed, 16 Dec 1992 18:52:54 EST

Page 659 of the Mathematica Book, second edition, gives an example of
the use of MatrixPower and MatrixExp:
m= {{0.4, 0.6}, {0.525, 0.475}}

MatrixPower[m,3]
{{0.465625, 0.534375}, {0.467578, {0.532422}}

Using the same matrix on my system (386/7 combination), I get very
different results from the example:

m = {{.4, .4}, {.525, .475}}
{{0.4, 0.4}, {0.525, 0.475}}

MatrixPower[m,3]
{{0.33175, 0.31425}, {0.412453, 0.390672}}

m.m.m
{{0.33175, 0.31425}, {0.412453, 0.390672}}

Yet Mathematica reports this result as accurate to 16 digits.

MatrixExp likewise gives a different result:

MatrixExp[m]
{{1.6553, 0.641593}, {0.842091, 1.7756}}

Why the discrepancy?  Is the manual wrong?  Is my system wrong?
What's up here?




-- 
Katherine (Williams) Derbyshire
kewms at kew.com

Riches are for spending.                        - Francis Bacon





  • Prev by Date: Re: What should Mma be, part II
  • Next by Date: re: Plotting a polygon
  • Previous by thread: Re:warping of density plots
  • Next by thread: Re: Problem with MatrixPower, v. 2.1