Independent/Dependent Phases 2: Classical Logistic Equation



 Science > Physics > Independent/Dependent Phases 2: Classical Logistic Equation

LINK TO THIS PAGE  


rating :  0   |  0


  Page 1 of 1

1

 
Topic: Science > Physics
User: "OsherD"
Date: 04 Oct 2005 01:29:51 PM
Object: Independent/Dependent Phases 2: Classical Logistic Equation

From Osher Doctorow


The revised Theorem now reads:
Theorem For the generalized standardized Logistic Differential
equation:
1) dy/dt = B(t)y(1 - y), B(t) > 0
the second derivative Dtt(y) > 0 iff B'(t)/B^2(t) > 2y - 1 or
equivalently:
2) y < (1/2) + B'(t)/(2B(t)^2)
The proof has already been given.
Let's apply this to the classical standardized Logistic Differential
equation:
3) dy/dt = ky(1 - y), 0 < = y < = 1, k > 0
Here B(t) = k > 0, B'(t) = 0 for all t, so (2) becomes:
4) y < 1/2
which is precisely the result from Dyt(y) = Dy(dy/dt) = 0.
Osher Doctorow
.

User: "OsherD"

Title: Re: Independent/Dependent Phases 2: Classical Logistic Equation 04 Oct 2005 01:45:44 PM

From Osher Doctorow

Are there any examples for which the conditions of the Theorem are not
met? Yes.
Let's consider the generalized standardized Logistic Differential
equation with B(t) = exp(-t):
1) dy/dt = exp(-t)y(1 - y)
Then B' (t) = -exp(-t) and B' (t)/B^2(t) = -exp(-t)/[exp(-t)]^2 =
-1/exp(-t) = -1/(1/exp(t)) = -exp(t). Then B' /B^2 = -exp(t) > 2y - 1
iff 2y < 1 - exp(t) or y < (1/2) - (1/2)exp(t). For t = 0, this says y
< 0. For t > 0, this says y < negative number, both of which violate
the definition of y.
When B(t) is replaced above by exp(t), the condition always holds, but
the acceleration Dtt(y) doesn't change from positive for any t > = 0.
However, asymptotically as t --> infinity, we get y < 1/2
approximately, as expected from Dyt.
Osher Doctorow
.
User: "OsherD"

Title: Re: Independent/Dependent Phases 2: Classical Logistic Equation 04 Oct 2005 01:52:29 PM

From Osher Doctorow


I meant to say "which is precisely the result from Dyt(y) = Dy(dy/dt),"
not the same thing "= 0", in two postings back in the line below
equation (4).
Osher Doctorow
.



  Page 1 of 1

1

 


Related Articles
Independent/Dependent Phases 27: Dyt vs Dtt Equation
Independent/Dependent Phases 6: Power pdf fX(x) = 1/2
Independent/Dependent Phases 26: A Beautiful Mind Via Saddle Points
Independent/Dependent Phases 33.1: Conjecture on Falicov-Kimball Lattice
Independent/Dependent Phases 34: Probable Correlation is Universe's Probable Influence on Intersections
Independent/Dependent Phases 3: What is pdf f(t) = 1/2 in Accelerating Universe?
Independent/Dependent Phases 10.2: Special Relativity 3
Independent/Dependent Phases 21: 0 and 1 as Phase Boundaries
Independent/Dependent Phases 17: P(A) < 1/2 Implies P(A-->B) > 1/2
Independent/Dependent Phases 22: 196 Papers on Long Memory/Hysteresis
Independent/Dependent Phases 10: Re-interpretation of f(x,y) = 1/2
Independent/Dependent Phases 19.2 Square Root of a Quadric
Independent/Dependent Phases 14.1: 1, 2, sqrt 3 in Generalized Isoperimetric
Independent/Dependent Phases 11: Univariate fX(x) = 1/2
Independent/Dependent Phases 19.3: Not Complete Concordance in Quadric
 

NEWER

pg.1612     pg.1232     pg.940     pg.716     pg.544     pg.412     pg.311     pg.234     pg.175     pg.130     pg.96     pg.70     pg.50     pg.35     pg.24     pg.16     pg.10     pg.6     pg.3     pg.1

OLDER