lec24
Markov Matrix
A=.1.2.7.01.990.3.3.4
properties
- All entryis ≥0
- All columns add to 1
- λ=1 is an eigen value
- all other eigen value ∣λi∣<1
steady state
uk=Aku0=c1λ1x1+c2λ2x2+⋯+cnλnxn
if λ1=1 , the steady state of uk ( k→∞ ) is c1x1 i.e. the x1 part of u0 .
why λ=1 is an eigenvalue?
A−1I=−.9.2.7.01−.010.3.3−.6
All columns in A−I adds up to 0 $\stackrel{?}{\longrightarrow} $ $ A-I$ is singular
Beacuse rows are dependent.
row1+row2+⋯+rown=0
i.e. (1,1,1) is in N((A−I)⊤)
Application of Marcov matrix: population
[ucalumass]0=[01000]
[ucalumass]k+1 =[.9.1.2.8][ucalumass]k
$.1 ofCaliforniapopulation \to$ Massachusetts
$.2 ofMassachusettspopulation \to$ California
find eigenvalue and eigenvectors:
λ=1,.7
u0=[01000]=31000[21]+32000[−11]
uk=c11k[21]+c2(.7)k[−11]
steady state
uk=31000[21]
Fourier Series
intro
xi are orthonormal basis.
v=x1q1+x2q2+⋯+xnqn
How to get xi ?
e.g.
q1⊤v=x1q1q1⊤+0+⋯+0=x1
i.e. (writing in the form of matrix)
⋮q1⋮⋯⋮qn⋮x1⋮xnQx=v=v
x=Q−1v=Q⊤v
Fourier Series
f(x)=a0+a1cosx+b1sinx+a2cos2x+b2sin2x+⋯
$q_i areorthogonal,and \sin , \cos$ are orthogonal funtions. How come?
orthogonal funtions
like inner product, f=cosx and g=sinx are orthogonal funtions means:
f⊤g=∫02πf(x)g(x)dx=21(sinx)202π=0
find the coefficient:
a1=cos⊤cos(x)f⊤cos(x)=∫02πcos(x)2dx∫02πf(x)cos(x)dx=π1∫02πf(x)cos(x)dx