Application To Matrix Algebra
Application To Matrix Algebra
Application To Matrix Algebra
1) MARKOV CHAINS
b(k)
x(k) = y(k)
m(k)
o(k)
1-db 0 0
A= 0
0 1-dy 0
0
0 0 1-dm
1-do
Then we have
x(k+1) = A. x(k)
1
1 0.9
P=
3.88
0.72
1.26
Assume a total tree population of 50,000 trees. Suppose the forest is newly
planted :-
50, 000
x(0) = 0
0
0
0.9 0 0 0.19
0
x(2) = 0 0.8 0 x(1) = 0.09
0
0 0 0.7 0.72
0.6 0
And after 45 years
0.19
x(3) = 50 000
0.09
0.72
0
0.1 0.2 0.3
After 15n years, where n = 1,2 …….the age distribution in the forest is given by
0.4
0.9 0 0 n
x(n) = 0
0 0.8 0
0
0 0.1 0 0.2 0.7 0.3
0.6 0.4
0.9 0 0 n
x(0) = 50 000 1
0
0
0 0.8 0
0
0
0
0 0 0.7
0.6
So the problem is to find the nth power of the matrix A.We have seen that
diagonalization technique may be helpful to solve this problem. Another problem
deals with the long term behavior of the sequence (n) when n gets large.
2) LINEAR ECONOMIC MODEL
Definition :
Quantity Supplied : amount of good that sellers are willing and able to sell
Supply curve : upward sloping line relating price to quantity supplied
Quantity demanded : amount of good that buyers are willing and able to buy
Demand curve : downward sloping line relating price to quantity demanded
Market equilibrium : quantity demand = quantity supply
Example :
Supply and demand model :
q = a0 – a1 (p+t) +a2y
q = B0 +B1p + B2w
where q is quantity, p is the pre-tax price, t is a lump sum tax, y is income and w
is weather and all the a and B parameters are positive constants
Behavioural equations :
q = a0 – a1 (p+t) +a2y
q = B0 +B1p + B2w
We could write this model as a three equation model with two behavioural
equation and one identity :
qd = a0 – a1 (p+t) +a2y
qs = B0 +B1p + B2w
qs = qd
q and p are endogenous variables , t, y and w are the exegonous variables the
variables which are given from outside the model.
Example of question :
Consider the following simple closed linear economic model (numerical example)
Y=C+I
C = 100 + 0.6Y
I = 20 + 0.15Y
The model can be written in matrix form as
1 -1 -1 Y 0
=
-0.6 1 0 C 100
-0.15 0 1 I 20
This takes the form A x = b, where x is the vector of variables, A is the matrix of
coefficients and b is the vector of constants. Premultiplying both sides of the
equation by the inverse of A will enable us to solve the system.
4 4 4
480
Y 2.4 3.4 0
C = 2.4 100 =
388
I 0.6 0.6 20
1.6
92
Now let’s look at the general version of this model, i.e. where the numerical
values of
the parameters and constants are not known, but are denoted by symbols.
Y =C+I
C = a + bY
I = e + fY
0 0.8 0
0
0 0 0.7
If we find the inverse of the A matrix of coefficients then we can solve for Y, C
and I
1 1 1
1 − (b + f ) 1 − (b + f ) 1 − (b + f )
Y
= 0
C 1 1 1 a
I 1 − (b + f ) 1 − (b + f ) 1 − (b + f )
e
1 1 1
1 − (b + f ) 1 − (b + f ) 1 − (b + f )
a+e
1
be + a(1-f)
= 1 −(b + f )
af + e (1-b)
e
Nt = seedling at time t
SEEDLING
2) Each year every individual either dies, remains in the same stages or
advances to the next stages. Young and adult trees produce seeds.
SEEDS
ps
qs
SEEDLING pn
fy qn
YOUNG TREE
py
fa qy
ADULT TREE
pa
Matrix Formulation :
St+1 = ps . St + 0 . Nt + fy .Yt + fa .At
Nt+1 = qs .St + pn .Nt + 0 . Yt + 0 .At
Yt+1 = 0. St + qn . Nt + py .Yt + 0 .At
At+1 = 0. St + 0 . Nt + qy . Yt + pa . At
ps 0 fy
Xt+1 = PXt
fa
Xt = St P= qs pn 0
Nt 0
0 qn py
Yt
0
At
0 0 qy
pa
This is the exmple of how matrix could be used in population growth. We could
see that not only can matrices provide an excellent way to keep track of changes
in age structured population but the application of linear algebra techniques
reveal certain biologically relevant characteristics.
Example :
Show that the law a + a holds for all a in F2.
Let's now choose a field of numbers to work with - it could be Q or it
could be F2, for example.
In order to be as nonspecific as possible, we'll call itK. Then we can think about
vectors with components in K. These are objects of the form (1, 0) or (0, 1, 0), or
(0, 1, 0, 1), etc.
More generally, a vector is an n-tuple (a1, a2, ……., an), and we call the set of all
such vectors Kn
The good thing about vectors is that they can be added. We do this by
adding their components:
(a1 , a2 , ….. , an) + (b1, b2 , …. , bn) = (a1 + b1, a2 + b2,….. , an + bn)