8/3/2019 N20122 Mathematical Economics II 1 141
1/141
1
Econ 416/616 Mathematical Economics II Lau
Lecture Notes Spring 2012
Review of Integration
Indefinite integral as anti-derivative:
1.
1
11 1
( )111 Note that ( 1)
1 1
a
aa a a
xd c
x a x dx c a a x xa dx a
Example:7
76 7 1 6
33
21 3 3 12
12 2 2 2
22
31 2 1 133 3 3
( )17Note that (7)
7 7
2( )
2 2 33Note that ( )3 3 3 2
2
3( )
3 3 22Note that ( )2 2 2 3
3
xd c
x x dx c x x
dx
d x c
x x dx c x c x xdx
d x cx
x dx c x c x xdx
3
2. 11 (ln ) 1
ln Note thatd x c
x dx dx x c x dx x
3.x x
e dx e c Example:2
22 2 2
( )12Note that (2)
2 2
x
x x x x
ed c
ee dx c e e
dx
Definite integral:
( )b
a
f x dx
( )f x
a b
8/3/2019 N20122 Mathematical Economics II 1 141
2/141
2
Example:22 2 2 2
1 1
22
1 1
22 5 10 55
1 1
(2) (1) 4 1 3
2 2 2 2 2 2
1
ln ln 2 ln1 ln 2
5 5
x
x
xxdx
dx xx
e e ee dx
33
0 0
1ln ln 3 ln 0dx x undefined
x
0 3
Integration by parts: udv uv vdu
Proof: duv udv vdu uv duv udv vdu
Example 1 Evaluate x xe dx
Let
x x x
u x du dx
dv e dx dv e dx v e
x x x x x xe dx xe e dx xe e c
Example 2 Evaluate ln x dx
Let
1lnu x du dx
x
dv dx dv dx v x
1ln (ln )( ) ( ) ln ln x dx x x x dx x x dx x x x c
x
unbounded
8/3/2019 N20122 Mathematical Economics II 1 141
3/141
3
Example 3 Evaluate 2 x x e dx
Let
2 2
x x x
u x du xdx
dv e dx dv e dx v e
2 2 2
( ) ( )(2 ) 2
x x x x x
x e dx x e e xdx x e xe dx
Let
x x x
u x du dx
dv e dx dv e dx v e
2 2 2 2
2 2
2 2 ( ) ( ) 2 ( ) ( )
2 2 ( 2 2)
x x x x x x x x x
x x x x
x e dx x e xe dx x e x e e dx x e x e e c
x e xe e c e x x c
Example 4 Evaluate xe cos x dx
Let
cos cos sin
x xu e du e dx
dv x dx dv x dx v x
sin sin x x xe cos x dx e x x e dx
Let
sin sin cos
x xu e du e dx
dv x dx dv x dx v x
sin sin sin ( cos ) ( cos )
sin cos cos
2 sin cos
1sin cos
2
x x x x x x
x x x x
x x x
x x x
e cos x dx e x x e dx e x e x x e dx
e cos x dx e x e x e x dx
e cos x dx e x e x
e cos x dx e x e x c
8/3/2019 N20122 Mathematical Economics II 1 141
4/141
4
First Order Liner Differential Equations
Example 1: '( ) 1 or 1dy
y tdt
1 ( )dy
dy dt dy dt y t t cdt
Note: When we differentiate ( ) with respect to y t t c t , we get '( ) 1y t .
Example 2: '( ) y t y
1 1ln
1ln ( ) wheret c cy tdy dy dy
y dt dt y t c e e y t Ce C edt y y
Note: When we differentiate ( ) t y t Ce with respect to t, we get '( ) t y t Ce y .
Example 3:2'( ) y t y
2 1
2 2
1( )
dy dy dy y dt dt y t c y t
dt t cy y
Note: 2 2( )dy
t c ydt
Example 4: 1'( ) y t y 21
( ) 2( )2
dy y ydy dt ydy dt t c y t t c
dt y
Note:1 2 1 1
2 2( ) 2( )
dy
dt yt c t c
Example 5: '( ) 2 (0) 1 y t y
2 2 2 ( ) 2
(0) 2(0) 1 1
( ) 2 1
dy dy dt dy dt y t t cdt
y c c
y t t
Note: 2 (0) 2(0) 1 1dy
ydt
8/3/2019 N20122 Mathematical Economics II 1 141
5/141
5
Example 6: '( ) 3 (0) 1 y t y y
13
13 3 ln 3 ( ) wherectdy dy
y dt y t c y t Ce C edt y
3(0)(0) 1 y Ce C
3
( )t
y t e
Note: 3 3(0)3 3 (0) 1tdy
e y y edt
Example 7: '( ) 3 (1) 4 y t y y 3( ) t y t Ce 3(1) 3(1) 4 4 y Ce C e 3 3 3 3( ) 4 4t t y t e e e
3 3 3(1) 3Note: 4(3) 3 (1) 4 4tdy
e y y edt
Example 8: '( ) ( ) y t Py t Q
[ ( )]'( ) ( ' )
PtPt Pt Pt Pt d e y t
e y t yPe e y Py e Qdt
[ ( )] ( ) ( ) ( ) (**)Pt Pt Pt Pt Pt Pt Pt Q Q
d e y t Qe dt de y t Qe dt e y t e C y t CeP P
Note: '( ) ( ) Pt Pt dy y t C P e CPedt
'( ) ( ) [ ]Pt Pt Pt Pt Q
y t Py t CPe P Ce CPe Q CPe QP
Example 9: '( ) 3 y t y
Using (**): ( 3 ) 30
3, 0 ( ) ( )3
t tP Q y t Ce y t Ce
Example 10: '( ) 3 4 (0) 1 y t y y
Using (**): ( 3) 34 4
3, 4 ( ) ( )3 3
t tP Q y t Ce y t Ce
3(0) 34 7 7 4(0) 1 ( )3 3 3 3
t y Ce C y t e
8/3/2019 N20122 Mathematical Economics II 1 141
6/141
6
Example 11: '( ) ( ) ( ) ( )y t P t y t Q t ( ) ( )
( ) ( ) (***)P t dt P t dt
y t e Q t e dt c
Example 12:1
'( ) 3 y t y t t
Using (***): 1( ) , ( ) 3P t Q t t t
1 1
ln 1 2 1 3 2( ) 3 3 ( ) ( 3 ) ( )dt dt
tt tc
y t e te dt c e t t dt c t t dt c t t c t t
Example13: 22
'( ) 6 y t y t t
Using (***): 22
( ) , ( ) 6P t Q t t t
2 2 5 3
2 2 ln 2 2 2 4 2
26 6( ) 6 6 ( ) ( 6 ) ( )5 5
dt dt tt t t t c y t e t e dt c e t t dt c t t dt c t ct
8/3/2019 N20122 Mathematical Economics II 1 141
7/141
7
Second Order Linear Differential Equation (Homogeneous)
"( ) '( ) ( ) 0 y t Ay t By t
Let ( )rt
y t ce be the solution of the differential equation.
2'( ) , "( )rt rt y t cre y t cr e
Substitute to the differential equation, we have2 20 ( ) 0rt rt rt rt r ce Acre Bce ce r Ar B
This leads to a characteristic equation: 2 0r Ar B
The characteristic roots:2
1 2
4,
2
A A Br r
Case 1 2 4A B
1 2
1 2( )r t r t
y t c e c e
Example: " 4 0y y
0, 4 orA B characteristic equation2
2
1 2
(0) (0) 4( 4)4 0 , 2, 2
2r r r
2 2
1 2( )t t
y t c e c e
Note:2 2
1 2
2 2
1 2
'( ) 2 2
"( ) 4 4
t t
t t
y t c e c e
y t c e c e
Example: 2 " 12 ' 10 0 y y y or " 6 ' 5 0 y y y
6, 5 orA B characteristic equation2
2
1 2
(6) (6) 4(1)(5) (6) 166 5 0 , 1, 5
2 2
r r r r
5
1 2( ) t t y t c e c e
Note:5 5
1 2 1 2'( ) 5 and "( ) 25t t t t y t c e c e y t c e c e
Hence5 5 5 5
1 2 1 2 1 2 1 1 1 2 2 2( 25 ) 6( 5 ) 5( ) ( 6 5 ) (25 30 5 ) 0t t t t t t t t c e c e c e c e c e c e c c c e c c c e
8/3/2019 N20122 Mathematical Economics II 1 141
8/141
8
Case 2 2 4A B (complex roots)
Suppose the characteristic roots 1 2,r r a bi , then 1 2( ) ( cos sin )at
y t e k bt k bt
Example: "( ) 4 '( ) 13 ( ) 0y t y t y t
4, 13 orA B characteristic equation2
2
1 2
( 4) ( 4) 4(13)4 13 0 , 2 3 , 2 3
2r r r r i i
2
1 2 1 2( ) ( cos sin ) ( cos3 sin3 )at t y t e k bt k bt e k t k t
Note:2
1 2( ) ( cos3 sin3 )t
y t e k t k t 2 2 2
1 2 1 2 2 1 1 2
2 22 1 1 2 2 1 1 2
2
2 1 1 2
'( ) [ 3 sin 3 3 cos3 ] ( cos3 sin 3 )(2 ) [(2 3 )sin 3 (2 3 )cos3 ]
"( ) [3(2 3 )cos3 3(2 3 )sin3 ] [(2 3 )sin3 (2 3 )cos3 ]2
{[(4 6 ) (6 9 )
t t t
t t
t
y t e k t k t k t k t e e k k t k k t
y t e k k t k k t k k t k k t e
e k k k k
2 1 1 2
2
2 1 2 1
]sin3 [(6 9 ) (4 6 )]cos3 }
{( 5 12 )sin 3 (12 5 )cos3 }t
t k k k k t
e k k t k k t
2 2
2 1 2 1 2 1 1 2
2
1 2
2 2 2 2 2 2
2 1 2 1 2 2 1 1 2 1
"( ) 4 '( ) 13 ( )
{( 5 12 )sin3 (12 5 )cos3 } 4 [(2 3 )sin 3 (2 3 )cos3 ]
13 ( cos3 sin 3 )
( 5 12 ) 4 (2 3 ) 13 sin 3 (12 5 ) 4 (2 3 ) 13 cos3
t t
t
t t t t t t
y t y t y t
e k k t k k t e k k t k k t
e k t k t
e k k e k k k e t e k k e k k k e t
2 2
2 1 2 1 2 2 1 1 2 1( 5 12 8 12 13 ) sin 3 (12 5 8 12 13 ) cos3 0
t tk k k k k e t k k k k k e t
sincos
cossin
d tt
dt
d tt
dt
8/3/2019 N20122 Mathematical Economics II 1 141
9/141
9
Case 3 2 4A B (equal roots)
1 2( ) ( )rt
y t e c c t
Example "( ) 4 '( ) 4 ( ) 0y t y t y t
4, 4 orA B characteristic equation2
2
1 2
( 4) (4) 4(4)4 4 0 , 2
2r r r r
2
1 2 1 2( ) ( ) ( )rt ty t e c c t e c c t
Note:2
1 2
2 2 2
2 1 2 2 1 2
2 2 2
2 1 2 2 1 2 2
( ) ( )
'( ) ( ) 2 ( ) ( 2 2 )
"( ) ( 2 2 )( 2 ) ( 2 ) (4 4 4 )
t
t t t
t t t
y t e c c t
y t e c e c c t c c c t e
y t c c c t e e c c c c t e
2 2 2
1 2 2 2 1 2 1 2
2
1 2 2 2 1 2 1 2
"( ) 4 '( ) 4 ( ) (4 4 4 ) 4 ( 2 2 ) 4 ( )
[(4 4 4 ) (4 8 8 ) (4 4 )] 0
t t t
t
y t y t y t c c c t e c c c t e e c c t
e c c c t c c c t c c t
Example "( ) 8 '( ) 16 ( ) 0y t y t y t
8, 16 orA B characteristic equation2
2
1 2
( 8) (8) 4(16)8 16 0 , 4
2r r r r
41 2 1 2( ) ( ) ( )
rt ty t e c c t e c c t
Note:4
1 2
4 4 4
2 1 2 2 1 2
4 4 4
2 1 2 2 1 2 2
( ) ( )
'( ) ( ) 4 ( ) ( 4 4 )
"( ) ( 4 4 )( 4 ) ( 4 ) (16 8 16 )
t
t t t
t t t
y t e c c t
y t e c e c c t c c c t e
y t c c c t e e c c c c t e
4 4 4
1 2 2 2 1 2 1 2
4
1 2 2 2 1 2 1 2
"( ) 8 '( ) 16 ( ) (16 8 16 ) 8 ( 4 4 ) 16 ( )
[(16 8 16 ) (8 32 32 ) (16 16 )] 0
t t t
t
y t y t y t c c c t e c c c t e e c c t
e c c c t c c c t c c t
8/3/2019 N20122 Mathematical Economics II 1 141
10/141
10
Second Order Linear Differential Equation (Non-homogeneous)"( ) '( ) ( ) ( ) y t Ay t By t f t
Solution: ( ) H P y t y y
i) if 1 2real roots and r r , 1 21 2( )r t r t
y t c e c eB
ii) if complex roots,1 2( ) ( cos sin )
at y t e k bt k bt
B
iii) if 1 2real roots and r r , 1 2( ) ( )rt
y t e c c t
B
Example: Find the Py of "( ) '( ) ( ) y t gy t hy t t
Let ( ) '( ) "( ) 0P P P y t at b y t a y t
LHS: "( ) '( ) ( ) 0 ( ) ( ) y t gy t hy t g a h at b aht ga bh
RHS: t
Comparing coefficients, we have
2
i)
ii)
ah ah
gga h ghga bh b
h h h
2( ) ( )P
h g y t at b t
h h
Example: Find the Py of2"( ) 4 '( ) 4 ( ) 2y t y t y t t
Let 2 '( ) 2 "( ) 2P P P y at bt c y t at b y t a
LHS:2 2"( ) 4 '( ) 4 ( ) (2 ) 4(2 ) 4( ) 4 (4 8 ) (2 4 4 ) y t y t y t a at b at bt c at b a t a b c
RHS: 2 2t
Comparing coefficients, we have
4 1, 4 8 0, 2 4 4 2a b a a b c 21 1 7 7
, , ( )4 2 8 4 2 8
P
t ta b c y t
solution for the homogeneous
particular solution for the non-homogeneous equation
Solving for Py when ( ) (constant)f t : ( ) '( ) 0 "( ) 0P P P y t y t y t B
This satisfies "( ) '( ) ( ) y t Ay t By t
When ( ) is a polynomialf t of degree n , then the Py will also be a polynomial of degree n .
8/3/2019 N20122 Mathematical Economics II 1 141
11/141
11
Example: Find the Py of "( ) '( ) ( )qt
y t y t y t pe
Let 2( ) '( ) "( )qt qt qt P P P y t Ae y t Aqe y t Aq e
LHS:2 2"( ) '( ) ( ) ( ) ( ) ( )qt qt qt qt y t y t y t Aq e Aqe Ae e Aq Aq A
RHS: qtpe
Comparing coefficients, we have
2
2 2( )
qt
P
p pe Aq Aq A p A y t
q q q q
Note that this only works when2 0q q .
The condition2 0q q means that q is not a solution of the characteristic equation
2 0r r , that is qte is not a solution of "( ) '( ) ( ) 0y t y t y t .
If q is a simple root of
2
0q q , we look for a constant B such thatqt
Bte is a solution.
If q is a double root of2 0q q , we look for a constant C such that 2 qtCt e is a solution.
Example: Find the Py of "( ) 4 '( ) 4 ( ) 2cos2 y t y t y t t
Let ( ) sin 2 cos2 '( ) 2 cos2 2 sin 2 "( ) 4 sin 2 4 cos2P P P
y t A t B t y t A t B t y t A t B t
LHS: "( ) 4 '( ) 4 ( ) [ 4 sin 2 4 cos2 ] 4[2 cos2 2 sin 2 ] 4[ sin 2 cos2 ]y t y t y t A t B t A t B t A t B t ( 4 8 4 )sin 2 ( 4 8 4 )cos2 8 sin 2 8 cos2A B A t B A B t B t A t
RHS: 2cos2t
Comparing coefficients, we have1 1
8 0 and 8 2 0 and ( ) sin 24 4
P B A B A y t t
When ( ) sin cos f t p rt q rt , then ( ) sin cosP y t A rt B rt
When ( ) qt f t pe , then ( ) qtP y t Ae
The corresponding homogeneous second order linear differential equation is
"( ) '( ) ( ) 0y t y t y t and the characteristic equation is 2 0r r
, , ,p q are given
8/3/2019 N20122 Mathematical Economics II 1 141
12/141
12
Note that these techniques for obtaining particular solutions also apply if ( )f t is a sum,
difference, or product of polynomials, exponential functions, or trigonometric functions.
Example:
If2 3 2 3
( ) ( 1) sin 2 ( ) ( ) sin 2 cos2t t
P f t t e t y t At Bt C e D t E t
8/3/2019 N20122 Mathematical Economics II 1 141
13/141
13
Example"( ) '( ) 0 y t Ay t
Let ( )rt
y t ce be the solution of the differential equation.
2'( ) , "( )rt rt y t cre y t cr e
Substitute to the differential equation, we have2 20 ( ) 0rt rt rt r ce Acre ce r Ar
This leads to a characteristic equation: 2 0r Ar
The characteristic roots:2
1 2, or 02 2
A A A Ar r A
(0)
1 2 1 2( )At t At
y t c e c e c e c
Note:2
1 1'( ) and "( ) At At y t c Ae y t c A e
Hence2
1 1 1 2"( ) '( ) ( ) ( ) 0( ) 0 At At At y t Ay t c A e A c Ae c e c
Example
Looking for the for "( ) '( )P
y y t Ay t
Conjecture: ( )P y t C
( ) '( ) 0 and "( ) 0P P P y t C y t y t Does not work
Conjecture: ( )P y t at
( ) '( ) and "( ) 0P P P y t at y t a y t
"( ) '( ) (0) ( )
( )P
y t Ay t A a aA
y t t A
Note:
'( ) and "( ) 0P P y t y t A
"( ) '( ) (0) ( )P P y t Ay t A
A
8/3/2019 N20122 Mathematical Economics II 1 141
14/141
14
Example
Looking for the for "( ) '( )P y y t Ay t t
Conjecture: ( )P y t at b
( ) '( ) and "( ) 0P P P
y t at b y t a y t
"( ) '( ) (0) ( )P P y t Ay t A a t
Conjecture: 2( )P y t at bt 2( ) '( ) 2 and "( ) 2
P P P y t at bt y t at b y t a
2
2
2
"( ) '( ) (2 ) (2 )
2 (2 )
22
2
2 22
( )2
P P
P
y t Ay t a A at b t
aAt a Ab t
aA aA
a Aa Ab b
A A A A
y t t t A A A
Note:
2'( ) + and "( )P t y t y t A A A A
2"( ) '( ) ( ) +
P P
t y t Ay t A t t
A A A A A A
Does not work!!
8/3/2019 N20122 Mathematical Economics II 1 141
15/141
15
Example
"( ) 0y t
Conjecture: ( ) y t at b
'( ) and "( ) 0 y t a y t
Hence the solution is ( ) for any , y t at b a b
Example
Looking for the for "( )P y y t
Conjecture: ( )P y t at b
( ) '( ) and "( ) 0P P P
y t at b y t a y t Does not work.
Conjecture: 2( )P y t at bt c
2
2
( ) '( ) 2 and "( ) 22
( ) for any ,2
P P P
P
y t at bt c y t at b y t a a
y t t bt c b c
Note:
'( ) and "( )P P y t t b y t
8/3/2019 N20122 Mathematical Economics II 1 141
16/141
8/3/2019 N20122 Mathematical Economics II 1 141
17/141
17
Steady State Equilibrium and Stable Equilibrium
Definition: Let '( ) ( )y t F y . If *y such that '( ) ( *) 0 y t F y , then *y is a steady state
equilibrium.
Definition: Let *y be a steady state equilibrium. It is a stable equilibrium if lim ( ) *t y t y .
Definition: Let ' ( ) y F y . If t does not appear explicitly on the right-hand side, it is called an
autonomous (or time-independent) differential equation.
Theorem: If ( *) 0F y and '( *) 0 *F y y is a stable equilibrium.
If ( *) 0F y and '( *) 0 *F y y is an unstable equilibrium.'y
stable equilibrium ( )F y
y
unstable equilibrium
( )'( )
dF yF y
dy
8/3/2019 N20122 Mathematical Economics II 1 141
18/141
18
Example:
Solows neoclassical economic growth model
Assumptions:
1. one output ( )Y t
2. linear homogenous production function ( ) [ ( ), ( )]Y t F K t L t
[ , ]
1Let [ ,1] ( ) per capita production function
Y F K L
Y KF y f k
L L L
a) marginal product of capital
1 1
'( )( ) 1
'( )
K
K
Y
y YL MP
MP f k K K L K L y f k
f kK K L
i) '( ) 0f k
ii)0 0
lim '( ) [Given , lim ]Kk K
f k L MP
iii) lim '( ) 0 [Given , lim 0]Kk K
f k L MP
iv)'( ) ''( )
''( ) 0 [ ]KMP f k f k
f kK K L
b) marginal product of labor
2
2
'( )( ) '( ) ( ) '( )
( )'( )
L L
YYL Y
y LLY f k K
L MP Y f k K MP f k kf k L L LL
y f k K f k
L L L
c) (0) 0f
d) ( )f
3. 0 1S sY s 4. 'K I 5. At equilibrium, I S
6.'
is exogenousL
n nL
8/3/2019 N20122 Mathematical Economics II 1 141
19/141
19
Model:
' '
( ) ( ) (1)
Y C I
Y C I
L L L
C K K C
f k f k L L L L
' ' ' 'ln ln ln
' ' '' ' (2)
K k K L K k k K L n
L k K L K
K K Kk k nk nk k nk
K L L
(1) and (2) ( ) 'C
f k k nk L
fund
' ( ) ( )
'( ) ( ) amental equation of neo-classical economic growth model
C S sY
k f k nk nk nk sf k nk L L L
k t sf k nk
nk
( )sf k
k
* steady state equilibriumk
' 0k
0 k
' 0k
8/3/2019 N20122 Mathematical Economics II 1 141
20/141
20
The stability of *k
@ *, * ( *)
* ( *)
k nk sf k
nk f k
s
( *) ( *) (0)
* * 0
n f k f k f
s k k
By the Mean Value Theorem,( *) (0)
(0, *) such that '( )* 0
f k f x k f x
k
Define so that (0,1)*
x
k and
( *) ( *) (0)'( *) (**)
* * 0
n f k f k f f k
s k k
.
Multiply (**) by *k * ( *) * '( *) * '( *)n
k f k k f k k f k s
* * '( *) '( *)n n
k k f k f k s s
'( *)n sf k
Note that ( ) '( ) ( )F k k t sf k nk fundamental equation
Hence '( ) '( ) and '( *) '( *) 0F k sf k n F k sf k n *k is a stable equilibrium.
Mean Value Theorem
Let ( )f x be differentiable on ( , )a b and continuous on [ , ]a b . Then there is at least one point c
where( ) ( )
'( )f b f a
f cb a
( ) y f x
y
x
a c b
(0) 0f
diminishing MPK
8/3/2019 N20122 Mathematical Economics II 1 141
21/141
21
System of first order ordinary differential equations
Definition
1 1 1 2
1 2
' ( , ,..., , )
...
' ( , ,..., , )
n
n n n
x f x x x t
x f x x x t
is a system of first order ordinary differential equations
Theorem (existence and uniqueness of solution)
If 1(...)
( , ... , , ) and ii n
j
f f x x t
x
are continuous, then for every point 0 0
0 1( , , ..., )
nt x x , a unique
solution ( )i ix t satisfying the system of differential equation and0
0( )
i ix t .
Definition If if does not depends (explicitly) on t, i.e.
1 1 1 2
1 2
' ( , ,..., )
...
' ( , ,..., )
n
n n n
x f x x x
x f x x x
is an autonomous system of first order ordinary differential equations
Definition Any point 1( *, ..., *)nx x is a critical point if
1 1 1 2
1 2
' ( *, *,..., *) 0
...
' ( *, *,..., *) 0
n
n n n
x f x x x
x f x x x
Definition 1( *, ..., *)nx x is an isolated critical point if a neighborhood of 1( *, ..., *)nx x
containing no other critical points.
Definition If if is linear in , 1,...,j x j n , then
1 1 1 2
1 2
' ( , ,..., )
...' ( , ,..., )
n
n n n
x f x x x
x f x x x
is a system oflinear ordinary differential equations.
8/3/2019 N20122 Mathematical Economics II 1 141
22/141
22
Properties of the solution of an autonomous system
For simplicity, we will consider a system of 2 equations.
' ( , )
' ( , )
x F x y
y G x y
(1) , and their derivatives are continuousF G
Lemma 1 If 1 2( ), ( ),x x t y y t t t t is a solution of (1), then for any real constant c , the
functions 1 1( ) ( ) and ( ) ( )x t x t c y t y t c are also solutions of (1).
Proof: By chain rule, we have 1 1'( ) '( ) and '( ) '( )x t x t c y t y t c .
1 1 1
1 1 1
'( ) '( ) [ ( ), ( )] [ ( ), ( )]
'( ) '( ) [ ( ), ( )] [ ( ), ( )]
x t x t c F x t c y t c F x t y t
y t y t c G x t c y t c G x t y t
1 1( ) and ( ) x t y t are solutions to (1).
Note: This lemma only applies for autonomous system.
Definition As t varies, a solution ( ), y( ) x t t of (1) describes parametrically a curve on the x y
plane. This curve is called a trajectory or a path.y
Example:
2
( )
( )
x t t
y t t
2
y x
x
Lemma 2 Through any point passes at most one path.
Remark: Note the distinction between solutions and trajectories of (1). A trajectory is a curve
that is represented by more than one solution.
Example:
2
( )
( )
x t t
y t t
and
2
4
( )
( )
x t t
y t t
are different solutions with the same trajectories.
Definition Let ( *, *)x y be an isolated critical point of (1). If [ ( ), ( )]C x t y t is a path of (1), then
we say that C APPROACHES ( *, *)x y as t if lim ( ) * and lim ( ) *t t
x t x y t y
.
Geometrically, if ( , )P x y is a point that traces out C in accordance with the
equations ( ) and ( )x x t y y t , then ( *, *) asP x y t .
8/3/2019 N20122 Mathematical Economics II 1 141
23/141
23
Definition If( ) *
lim( ) *t
y t y
x t x
exists, or if it becomes either positively infinite or negatively infinite as
t , then we say that C ENTERS the critical point ( *, *x y ) as t .
Note: The quotient is the slope of the line joining ( *, *x y ) and the point ( ( ), ( ))P x t y t .
Note: This property is a property of the path C, and it does not depend on which solution isused to represent the path.
Definition A critical point is said to be STABLE if for each 0R , r R such that every path
which is inside the circle 2 2 2 0( *) ( *) @ x x y y r t t will remain inside the circle2 2 2
0( *) ( *) for x x y y R t t .
Note: Loosely speaking, a critical point is stable if all paths that get sufficiently close to the
critical point will stay close to the critical point.
Definition A critical point is ASYMPTOTICALLY STABLE if it is stable and a circle2 2 2
0( *) ( *) x x y y r such that every path which is inside this circle 0@t t
will approach the critical point as t .
Definition A critical point that is not stable is UNSTABLE.
Solving a pair of linear differential equation/ Phase diagram of the solution
1 1
2 2
'
'
x a x b y
y a x b y
(1) homogeneous system
Theorem A If the homogeneous system has 2 solutions 1
1
( )
( )
x x t
y y t
and 2
2
( )
( )
x x t
y y t
, then
1 1 2 2
1 1 2 2
( ) ( )
( ) ( )
x c x t c x t
y c y t c y t
(2) is also a solution for the system for any constant 1 2andc c .
Theorem B If the 2 solutions in Theorem A has a Wronskian1 2
1 2
( ) ( )( )
( ) ( )
x t x t W t
y t y t
that does not
vanish, then (2) is the general solution of the system.
Example:' 4
' 2
x x y
y x y
has 2 solutions
3
3
t
t
x e
y e
and
2
22
t
t
x e
y e
.
1 2
1 2
3 2
5
3 2
( ) ( )( )
( ) ( )
02
t t
t
t t
x t x t W t
y t y t
e ee
e e
8/3/2019 N20122 Mathematical Economics II 1 141
24/141
24
Tryrt
rt
x Ae
y Be
,A B are constants to be determined
Substitute this into the system, 1 1
2 2
'
'
x a x b y
y a x b y
1 1
2 2
'
'
rt rt rt
rt rt rt
x rAe a Ae b Be
y rBe a Ae b Be
1 1
2 2
rA a A b B
rB a A b B
1 1
2 2
( ) 0
( ) 0
a r A b B
a A b r B
(3)
Note that 0A B are the trivial solution of the system.
Note that (3) has a non-trivial solution if 1 1
2 2
0a r b
a b r
1 1 2
1 2 1 2 1 2 1 2 2 1
2 2
0 ( )( ) 0 ( ) 0a r b
a r b r b a r a b r a b a ba b r
2
1 2 1 2 1 2 2 1( ) ( ) 4( )
2
a b a b a b a br
Solution:1
1
1 1
1 1
( )
( )
r t
r t
x t A e
y t B e
and
2
2
2 2
2 2
( )
( )
r t
r t
x t A e
y t B e
Solving the system 1 1
2 2
'
'
x a x b y
y a x b y
Note that
1
1 12
1 1 2 2
2 2
0
00 if 0
b
a r bb r
a r b a b r
a b r
characteristic equation
8/3/2019 N20122 Mathematical Economics II 1 141
25/141
25
Case 1: Distinct real roots
General solution:1 2
1 2
1 1 2 2
1 1 2 2
( )
( )
r t r t
r t r t
x t c A e c A e
y t c B e c B e
Example:'
' 4 2
x x y
y x y
(1 ) 0 1 1 0
4 ( 2 ) 0 4 2 0
r A B r A
A r B r B
Characteristic equation:2(1 )( 2 ) (1)(4) 0 2 2 4 0r r r r r
2
1 26 0 ( 3)( 2) 0 3, 2r r r r r r
1 3r :[1 ( 3)] 0 4 0
4 [ 2 ( 3)] 0 4 0
A B A B
A B A B
3
1
3
1
( )
( ) 4
t
t
x t e
y t e
22r :
[1 (2)] 0 0
4 [ 2 (2)] 0 4 4 0
A B A B
A B A B
2
2
2
2
( )
( )
t
t
x t e
y t e
General solution:3 2
1 2
3 2
1 2
( )
( ) 4
t t
t t
x t c e c e
y t c e c e
Suppose (0) 1, (0) 2x y .
3(0) 2(0)
1 2 1 2
3(0) 2(0)1 2 1 2
1 (0)
2 (0) 4 4
x c e c e c c
y c e c e c c
1 2
1 6and
5 5c c
3 2
3 2
1 6( )
5 5
4 6( )5 5
t t
t t
x t e e
y t e e
A simple non-trivial solution is
1 and 4A B
A simple non-trivial solution is
1 and 1A B
8/3/2019 N20122 Mathematical Economics II 1 141
26/141
26
Case 1A Real and distinct, same sign: Node Assume 1 2 0r r
General solution:1 2
1 2
1 1 2 2
1 1 2 2
( )
( )
r t r t
r t r t
x t c A e c A e
y t c B e c B e
Note: i) critical point (0,0) ii) 1 2lim lim 0
r t r t
t te e
iii) lim ( ) lim ( ) 0t t
x t y t
[i.e. ( ( ( ), ( )) (0,0) as ] x t y t t
iv) 1 2
1 2
( the roots are distinct)B B
A A
When2 2
22
2 2 2 21
2 22 2
( ) ( )0
( )( )
r t r t
r tr t
x t c A e B e By tc
x t A e Ay t c B e
When1 1
11
1 1 1 12
1 11 1
( ) ( )0
( )( )
r t r t
r tr t
x t c A e B e By tc
x t A e Ay t c B e
When
1 2
1 2 1 2
1 21 21 2
( )1 12
1 1 2 2 1 1 2 2 21 2
( )1 11 1 2 21 1 2 22
2
( ) ( )0, 0
( )( )
r r t
r t r t r t r t
r t r t r t r t r r t
c Be B
x t c A e c A e c B e c B e cy tc c
c Ax t c A e c A ey t c B e c B e e Ac
1 2
1 2
( )1 12
2 2
( )1 1 2
22
lim
r r t
t r r t
c Be B
c B
c A A
e Ac
as 1 2 0r r
Also
2 1
1 2
1 2
2 1
( )2 21
1 1 2 2 1
( )2 21 1 2 21
1
( )
( )
r r t
r t r t
r t r t r r t
c BB e
c B e c B e cy t
c Ax t c A e c A eA e
c
2 1
2 1
( )2 21
1 1
( )2 2 11
1
lim
r r t
t r r t
c BB e
c B
c A AA e
c
as 2 1 0r r
y
1
1
B
A
Node: Asymptotically stable
x
2
2
B
A
1 2andc c are arbitrary constants
8/3/2019 N20122 Mathematical Economics II 1 141
27/141
8/3/2019 N20122 Mathematical Economics II 1 141
28/141
28
Case 2 Equal real roots: Node 1 2r r r
General solution: 1 2 1 2
1 2 1 2
( ) ( )
( ) ( )
rt rt
rt rt
x t c Ae c A A t e
y t c Be c B B t e
Note that21 2 1 2 1 2 2 1( ) ( ) 4( )
2
a b a b a b a br
2
1 2 1 2 1 2 2 1( ) 4( ) 0r r r a b a b a b
Case 2A 1 2 2 10 and 0a b a a b 2 2 2
1 2 1 2 2 1 1 2 1 2 1 2[( ) 4( ) ( ) 4 ( ) 0]a b a b a b a b a b a b
1
2
' ( )
' ( )
rt
rt
x ax x t c e
y ay x t c e
If 0r , then lim ( ) 0 and lim ( ) 0t t
x t y t
If 0r , then lim ( ) 0 and lim ( ) 0t t
x t y t
1 2
12
( ) ( )
( )( )
rt
rt
x t c e cy t
x t cy t c e
y 2
1
c
c 0r
Node: Asymptotically stable
x
2
2
B
A
1 2andc c are arbitrary constants
8/3/2019 N20122 Mathematical Economics II 1 141
29/141
29
Case 2B Any other combination: Node
General solution: 1 2 1 2
1 2 1 2
( ) ( )
( ) ( )
rt rt
rt rt
x t c Ae c A A t e
y t c Be c B B t e
If 0, lim ( ) 0 and lim ( ) 0t t
r x t y t
[Note that the term t is dominated by the term rte .]
If 0, lim ( ) 0 and lim ( ) 0t t
r x t y t .
When 12
1
( )0
( )
rt
rt
c Be y t Bc
x t c Ae A
When 2 1 2 1 212 1 2 1 2
( )( )0
( ) ( )
rt
rt
c B B t e B B t y tc
x t c A A t e A A t
12
1 2 2
11 2 22
lim limt t
BB
B B t Bt
A A A t AAt
When
1 12 2
1 2 1 21 2
1 11 2 1 22 2
( )( )( )
0, 0( ) ( )
( )
rt rt
rt rt
c B Bc B
c Be c B B t ey t t tc cc A Ax t c Ae c A A t e
c At t
1 12 2
2
1 1 22 2
( )
lim
( )t
c B Bc B
Bt tc A A A
c At t
y
0r Node: Asymptotically stable
x
B
A
2
2
B
A
8/3/2019 N20122 Mathematical Economics II 1 141
30/141
30
Example:
' 3 4
'
x x y
y x y
In order to have a nontrivial solution forA
B
,3 4
0
1 1
r
r
2 2(3 )( 1 ) ( 4)(1) 0 2 1 0 ( 1) 0r r r r r real and equal roots
(3 1) 4 0 2 4 0(1)
(1 1) 0 2 0
A B A B
A B A B
A simple non-trivial solution of this system is 2, 1A B so that( ) 2
( )
t
t
x t e
y t e
A second linearly independent solution:
1 2 1 2 2 1 2 2
1 2 1 2 2 1 2 2
( ) ( ) '( ) ( ) ( )
( ) ( ) '( ) ( ) ( )
t t t t
t t t t
x t A A t e x t A A t e e A A A t A e
y t B B t e y t B B t e e B B B t B e
system 1 2 2 1 2 1 2
1 2 2 1 2 1 2
( ) 3( ) 4( )
( ) ( ) ( )
t t t
t t t
A A t A e A A t e B B t e
B B t B e A A t e B B t e
2 2 1 2 11 2 2 1 2 1 2
2 2 1 1 21 2 2 1 2 1 2
(2 4 ) (2 4 ) 0( ) 3( ) 4( )
( 2 ) ( 2 ) 0( ) ( ) ( )
t t t
t t t
A B t A A BA A t A e A A t e B B t e
A B t A B BB B t B e A A t e B B t e
2 2 1 2 1
2 2 1 1 2
2 4 0 2 4 0
2 0 2 0
A B A A B
A B A B B
The 2 equations on the left 2 22, 1A B .
The 2 equations on the right 1 1 1 1 1 12 4 2 and 2 1 1 and 0 A B A B A B
( ) (1 2 )(2)
( )
t
t
x t t e
y t te
1 2
1 2
( ) 2 (1 2 )general solution:
( )
t t
t t
x t c e c t e
y t c e c te
rt
rt
x Ae
y Be
8/3/2019 N20122 Mathematical Economics II 1 141
31/141
31
Case 3 Distinct complex roots: Spirals
1 1
2 2
'
'
x a x b y
y a x b y
1 2,r r a bi
General solution: 1 1 2 2 1 2
1 1 2 2 1 2
( ) [ ( cos sin ) ( sin cos )]( ) [ ( cos sin ) ( sin cos )]
at
at x t e c A bt A bt c A bt A bt y t e c B bt B bt c B bt B bt
If 1 2 0a a b , then lim ( ) 0 and lim ( ) 0t t
x t y t
, i.e. ( ( ), ( )) (0,0) as x t y t t , i.e. (0,0) is
asymptotically stable.
However, ( ) and ( ) x t y t change sign infinitely often as t , all paths must spiral to the criticalpoint.
y
0a Spiral: Asymptotically stable
x
Case 4: Pure imaginary roots: Center
1 2,r r bi
General solution:1 1 2 2 1 2
1 1 2 2 1 2
( ) ( cos sin ) ( sin cos )
( ) ( cos sin ) ( sin cos )
x t c A bt A bt c A bt A bt
y t c B bt B bt c B bt B bt
y
Center:stable but not Asymptotically stable
x
8/3/2019 N20122 Mathematical Economics II 1 141
32/141
32
Non-homogeneous system
1 1 1
2 2 2
'
'
x a x b y c
y a x b y c
Solving this system 1 1 1
2 2 2
* * 0
* * 0
a x b y c
a x b y c
will give us the critical point ( *, *)x y .
Define ( ) ( ) * and ( ) ( ) *x t x t x y t y t y [ '( ) '( ) and '( ) '( )x t x t y t y t ]
The original system can be written as1 1
2 2
'
'
x a x b y
y a x b y
The properties of the critical point of the non-homogeneous system are the same as those of the critical
point for the corresponding homogeneous system.
1 1 1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2 2 2
' ' ( *) ( *) * *
' ' ( *) ( *) * *
x a x b y x a x x b y y a x b y a x b y a x b y c
y a x b y y a x x b y y a x b y a x b y a x b y c
8/3/2019 N20122 Mathematical Economics II 1 141
33/141
33
Example:
' 3
' 4 2
x x y
y x y
( 3 ) 0
4 ( 2 ) 0
r A B
A r B
In order to have a nontrivial solution forA
B
,3 1
04 2
r
r
2 2
2
( 3 )( 2 ) (4)(1) 0 5 6 4 0 5 2 0
5 5 4(1)(2) 5 17
2 2
r r r r r r
r
1
5 17
2r
1 1 1 1
1 1 1 1
5 17 1 17( 3 ) 0 0
2 2
5 17 1 174 (2 ) 0 4 0
2 2
A B A B
A B A B
1 1 1 1
1 11 1
1
1
1 17 1 17 1 17 1 170 4 0
2 2 2 2
1 171 17 4 04 0
224 8
01 17 1 17
2
A B A B
A BA B
B
A
2
5 17
2r
2 2 2 2
2 2 2 2
5 17 1 17( 3 ) 0 0
2 2
5 17 1 174 (2 ) 0 4 0
2 2
A B A B
A B A B
2 2 2 2
2 21 2
2
2
1 17 1 17 1 17 1 170 4 02 2 2 2
1 171 174 04 0
22
4 80
1 17 1 17
2
A B A B
A BA B
B
A
rt
rt
x Ae
y Be
Both roots are real and
negative node
8/3/2019 N20122 Mathematical Economics II 1 141
34/141
34
y
' 0x
1
1
8 2.561 17
BA
' 0y Node: Asymptotically stable
x
2
2
81.56
1 17
B
A
General solution:1 2
1 2
1 1 2 2
1 1 2 2
( )
( )
r t r t
r t r t
x t c A e c A e
y t c B e c B e
Also
2 1
1 2
1 22 1
( )2 21
1 1 2 2 1
( )2 21 1 2 21
1
( )
( )
r r t
r t r t
r t r t r r t
c BB e
c B e c B e cy t
c Ax t c A e c A eA e
c
2 1
2 1
( )2 21
1 1
( )2 2 11
1
lim
r r t
t r r t
c BB e
c B
c A AA e
c
as 2 1 0r r
Suppose (0) 1 (0)x y
1 1 2 20 0
1 1 2 2
0 01 1 2 21 1 2 2
11 (0)
8 811 (0)
1 17 1 17
c A c Ax c A e c A e
c A c Ay c B e c B e
1 2
1
1 2 2
18 8
11 17 1 17
A Ac
A A c
2
2 2
1
1 2
1 2 1
1 2
1
8 8 81 1 1
1 17 1 17 1 17
8 8 8 8
8 8 1 17 1 17 1 17 1 17
1 17 1 17
A
A A
cA A
A A A
A A
8/3/2019 N20122 Mathematical Economics II 1 141
35/141
35
1
1 1
2
1 2
1 2 2
1 2
1
8 8 81 1 1
1 17 1 17 1 17
8 8 8 8
8 8 1 17 1 17 1 17 1 171 17 1 17
A
A A
cA A
A A A
A A
1 2
1
1 2
1 2
1
1
8 81 1
1 17 1 17( )
8 8 8 8
1 17 1 17 1 17 1 17
8 81 1
81 17 1 17( )
8 8 1 17
1 17 1 17
r t r t
r t
x t A e A e
A A
y t A e
A
2
2
2
8
8 8 1 17
1 17 1 17
r tA e
A
1 2
1
8 81 1
1 17 1 17( )
8 8 8 8
1 17 1 17 1 17 1 17
8 81 1
81 17 1 17
( ) 8 8 8 81 17
1 17 1 17 1 17
r t r t
r t
x t e e
y t e
28
1 17
1 17
r t
e
8/3/2019 N20122 Mathematical Economics II 1 141
36/141
36
Example:
' 2 3
'
x x y
y x y
(2 ) 3 0
(1 ) 0
r A B
A r B
In order to have a nontrivial solution forA
B
,2 3
01 1
r
r
2 2
2
(2 )(1 ) (3)(1) 0 3 2 3 0 3 1 0
( 3) ( 3) 4(1)( 1) 3 13
2 2
r r r r r r
r
1
3 13
2r
1 1 1 1
1 1 1 1
3 13 1 13(2 ) 3 0 3 0
2 2
3 13 1 13(1 ) 0 0
2 2
A B A B
A B A B
1 1 1 1
1 11 1
1
1
1 13 1 13 1 13 3 3 133 0 3 0
2 2 2 2
1 131 13 00
223 2
0.4343 03 3 13 1 13
2
A B A B
A BA B
B
A
2
3 13
2r
2 2 2 2
2 2 2 2
3 13 1 13(2 ) 3 0 3 0
2 2
3 13 1 13(1 ) 0 0
2 2
A B A B
A B A B
2 2 2 2
2 22 2
2
2
1 13 1 13 1 13 3 3 13
3 0 3 02 2 2 2
1 131 13 0022
3 20.768 0
3 3 13 1 13
2
A B A B
A BA B
B
A
The roots are real and of opposite signs saddle point
rt
rt
x Ae
y Be
8/3/2019 N20122 Mathematical Economics II 1 141
37/141
37
y
1
1
20.4343
1 13
B
A
Saddle Point
x
2
2
20.768
1 13
B
A
When2
2
2 2 21
22 2
( ) ( )0 , lim ( ) 0, lim ( ) 0, lim ( ) lim ( )
( )( )
r t
r t t t t t
x t c A e By tc x t y t x t y t
x t Ay t c B e
When1
1
1 1 12
11 1
( ) ( )0 , lim ( ) lim ( ) , lim ( ) 0, lim ( ) 0
( )( )
r t
r t t t t t
x t c A e By tc x t y t x t y t
x t Ay t c B e
When1 2
1 2
1 1 2 2
1 2
1 1 2 2
( )0, 0 lim ( ) lim ( )
( )
r t r t
r t r t t t
x t c A e c A ec c x t y t
y t c B e c B e
1 2
1 2
1 21 2
( )1 12
1 1 2 2 2
( )1 11 1 2 22
2
( )( )
r r t
r t r t
r t r t r r t
c Be B
c B e c B e cy tc Ax t c A e c A e
e Ac
1 2
1 2
( )1 12
2 2
( )1 1 22
2
lim
r r t
t r r t
c Be B
c Bc A A
e Ac
as 1 2 0r r
Also
2 1
1 2
1 22 1
( )2 21
1 1 2 2 1
( )2 21 1 2 21
1
( )
( )
r r t
r t r t
r t r t r r t
c BB e
c B e c B e cy t
c Ax t c A e c A eA e
c
2 1
2 1
( )2 21
1 1
( )2 2 11
1
lim
r r t
t r r t
c BB e
c B
c A AA e
c
as 2 1 0r r
Suppose (0) 1 (0)x y
1 1 2 20 0
1 1 2 2
0 01 1 2 21 1 2 2
11 (0)
2 211 (0)
1 13 1 13
c A c A x c A e c A e
c A c Ay c B e c B e
1 20r r
8/3/2019 N20122 Mathematical Economics II 1 141
38/141
38
1 2
1
1 2 2
12 2
11 13 1 13
A Ac
A A c
2
2 2
1
1 2
1 2 1
1 2
1
2 2 21 1 11 13 1 13 1 13
2 2 2 2
2 2 1 13 1 13 1 13 1 13
1 13 1 13
A
A A
cA A
A A A
A A
1
1 1
2
1 2
1 2 2
1 2
1
2 2 21 1 1
1 13 1 13 1 13
2 2 2 2
2 2 1 13 1 13 1 13 1 13
1 13 1 13
A
A A
cA A
A A AA A
1 2
1
1 2
1 2
1
1
2 21 1
1 13 1 13( )
2 2 2 2
1 13 1 13 1 13 1 13
2 21 1
21 13 1 13( )
2 2 2 21 13
1 13 1 13 1 13 1 13
r t r t
r t
x t A e A e
A A
y t A e
A
22
2
2
1 13r tA e
A
1 2
1
2 21 1
1 13 1 13( )
2 2 2 2
1 13 1 13 1 13 1 13
2 21 1
2 21 13 1 13
( ) 2 2 2 21 13 1 13
1 13 1 13 1 13 1 13
r t r t
r t
x t e e
y t e
2r t
e
8/3/2019 N20122 Mathematical Economics II 1 141
39/141
39
Suppose1 13
(0) 1, (0)6
x y
0 01 1 2 21 1 2 2
0 01 1 2 21 1 2 2
11 (0)
1 13 2 21 13 (0)6 1 13 1 136
c A c A x c A e c A e
c A c Ay c B e c B e
1 2
1
1 2 2
1
2 2 1 13
1 13 1 13 6
A Ac
A A c
2
22
1
1 2
1 2 1
1 2
1
1
2 1 13 2 1 13 1 131 13 2
6 61 13 1 13 1 136 1 132 2 2 2
2 2 1 13 1 13 1 13 1 13
1 13 1 13
2(1 13) 1 13 2(1 13) 1 13
1 13 6 12 6
2 2 2 2
1 13 1 13 1 13 1 13
A
AA
cA A
A A A
A A
A A
1 1
(1 13) 1 13
6 6 02 2
1 13 1 13A
1
11
2
1 2
1 2 2
1 2
2
11 13 2 1 13 22 1 13
6 61 13 1 1361 13
2 2 2 2
2 2 1 13 1 13 1 13 1 13
1 13 1 13
( 1 13) 1 13 2 6 (1 2 13 13) 12
6 61 13 1 13 6(1 13)
2 2 2
1 13 1 13 1 1
A
AA
cA A
A A A
A A
A
2 2
2
2 13 26
6(1 13)
2 2 2
3 1 13 1 13 1 13
13 13
2 23(1 13)
1 13 1 13
A A
A
8/3/2019 N20122 Mathematical Economics II 1 141
40/141
40
2 2
2
2
2
2
2
13 13 13 13( ) =
2 2 2 23(1 13) 3(1 13)
1 13 1 13 1 13 1 13
13 13 2 13 13 2
( ) 2 2 2 21 13 13(1 13) 3(1 13)
1 13 1 13 1 13 1 13
r t r t
r t
x t A e e
A
y t A eA
2
13
r
e
Note that as , ( ) 0 and ( ) 0t x t y t .
In general, so long as the given point fulfill the condition( ) 1 13
0.768( ) 6
y t
x t
, then the path will
converge to the critical point.
The critical point is a SADDLE POINT.
8/3/2019 N20122 Mathematical Economics II 1 141
41/141
41
Example:
' 5 2
' 17 5
x x y
y x y
(5 ) 2 0
17 ( 5 ) 0
r A B
A r B
In order to have a nontrivial solution forA
B
,5 2
017 5
r
r
2 2(5 )( 5 ) (2)( 17) 0 25 34 0 9 0
3
r r r r
r i
General solution:
1 1 2 2 1 2
1 1 2 2 1 2
( ) ( cos3 sin 3 ) ( sin 3 cos3 )
( ) ( cos3 sin3 ) ( sin 3 cos3 )
x t c A t A t c A t A t
y t c B t B t c B t B t
The roots are complex roots center
rt
rt
x Ae
y Be
8/3/2019 N20122 Mathematical Economics II 1 141
42/141
42
Taylor Expansion2"( *)( *) ( )( *)
( ) ( *) '( *)( *) ... ( , *)2! !
n n f x x x f a x x
f x f x f x x x a x xn
2 2
( , ) ( *, *) ( *, *)( *) ( *, *)( *)
( *, *)( *) 2 ( *, *)( *)( *) ( *, *)( *)...
2!
x y
xx xy yy
f x y f x y f x y x x f x y y y
f x y x x f x y x x y y f x y y y
Example: Expand ( ) around 0x f x e x
( ) (0) 1
'( ) '(0) 1
''( ) ''(0) 1
...
x
x
x
f x e f
f x e f
f x e f
2 3
2 3 4
''(0) '''(0)( ) (0) '(0)( 0) ( 0) ( 0) ...2! 3!
1 ...2! 3! 4!
f f f x f f x x x
x x xx
Example: Expand ( ) ln(1 ) around 0 f x x x .
1
2
( ) ln(1 ) (0) 0
'( ) (1 ) '(0) 1
''( ) 1(1 )
f x x f
f x x f
f x x
3 3
4 4 4
''(0) 1
'''( ) ( 2)(1 ) =(2!)(1 ) '''(0) 2!
( ) 2( 3)(1 ) = (3!)(1 )
f
f x x x f
f x x x
4
5 5 5 5
(0) 3!
( ) (2)(3)( 4)(1 ) (4!)(1 ) (0) 4!
...
f
f x x x f
2 3
2 3 4 5 2 3 4 5 6
''(0) '''(0)( ) (0) '(0)( 0) ( 0) ( 0) ...
2! 3!
( 1) (2!) ( 3!) (4!)( ) 0 ... ...
2! 3! 4! 5! 2 3 4 5 6
f f f x f f x x x
x x x x x x x x x f x x x
8/3/2019 N20122 Mathematical Economics II 1 141
43/141
43
Phase diagram
Shell Model of Inventive Activity and Capital Accumulation
( ) [ ( ), ( ), ( )]Y t F A t K t L t
( ) : stock of technical knowledge
( ) : capital stock
( ) : labor
A t
K t
L t
: output allocated for inventive activity
(1 )(1 ) : output allocated for consumption
(1 ) : output allocated for capital accumulation
Y
s Y
sY
success rate depreciation
'( ) ( ) ( )
'( ) (1 ) ( ) ( )
A t Y t A t
K t s Y t K t
Let ( , ) where ( , ) is homogenous to degree 1Y AF K L F K L ( ) ( ) 0, ' 0, " 0, (0) 0, ( ) , '(0) , '( ) 0 y Af k Y ALf k f f f f f f f
Let' ' ' ' ' '
1 so that ( ); ; ' 'Y k K L K K K
L Y Af k y Y k k K L k K L K K L
' ( ) (1)
' (1 ) (1 ) ( ) (2)
A Y A Af k A
k s y k s Af k k
Constructing the ' 0A curve
1' ( ) 0 ( ) 0 ( ) ( )S A Af k A f k f k k f
A
' 0A
I J
1( )
Sk f
k
Consider the points ( , ) and ( , ) I k A J k A I I J J
.
Since
i) ( ) ( ) f k f k I J
ii) ( ) 0Jf k
' ( ) 0I A f k
Since andI Jare arbitrary ' 0A on the left of the' 0A curve.
8/3/2019 N20122 Mathematical Economics II 1 141
44/141
44
Constructing the ' 0k curve
' (1 ) ( ) 0(1 ) ( )
kk s Af k k A
s f k
Differentiating
2 2
' 0
[ (1 ) ( )] ( )[ (1 ) '( )] [ ( ) '( )]with respect to 0[ (1 ) ( )] (1 )[ ( )]k
dA s f k k s f k f k kf k A k dk s f k s f k
The ' 0k is upward sloping.
A ' 0k
I
J
k
A
' 0A ' 0k
SA saddle point
k
Sk
Consider the points ( , ) and ( , ) I k A J k A I I J J
.
Since and I J I J A A k k
(1 ) ( ) (1 ) ( ) 0 I I I J J J s A f k k s A f k k
Since andI Jare arbitrary ' 0k for the pointswhich lie above the ' 0k curve.
Similarly, ' 0k all the points which lie below the' 0k curve.
8/3/2019 N20122 Mathematical Economics II 1 141
45/141
45
Stability
Take a Taylor series expansion around ( , )S Sk A .
' ' ' '
' '( , ) ( ) ( ) ( ) ( )
' ' ' '' '( , ) ( ) ( ) ( ) ( )
S S S S
S S S S
S S S S
S S S S
S S S S S S A A A A A A A Ak k k k k k k k
S S S S S S A A A A A A A Ak k k k k k k k
A A A A
A A k A A A k k A A k k A k A k
k k k k k k k A A A k k A A k k
A k A k
1
'( ) 0
S
S
SA Ak k
A f k a
A
1
''( ) 0
S
S
S SA Ak k
A A f k b
k
2' (1 ) ( ) 0S
S
SA Ak k
k s f k aA
2
'(1 ) '( ) (sign unknown)
S
S
S SA Ak k
ks A f k b
k
2 2
1 2 1 2 1 2 2 1 2 2 2 1( ) ( ) 4( ) 4,
1 2 2 2
a b a b a b a b b b a br r
2 2 2 2 22 2 2 1 2 2 2 1
1 2 2 1
( 4 ) ( 4 )0
4 4
b b a b b b a br r a b
1 2,r r are real and opposite sign
saddle point
8/3/2019 N20122 Mathematical Economics II 1 141
46/141
46
Phase Diagram of'( ) '( ')
" , convex functions, " 0, " 0"( ')
g x rf x x f g f g
f x
We will construct a phase diagram with and 'x x as the 2 variables.
Constructing ' 0x 'x
' 0x x
Constructing " 0x " 0 '( ) '( ') 0 x g x rf x
Total differential "( ) "( ') ' 0g x dx rf x dx
' 0
' "( )0
"( ')x
dx g x
dx rf x
'x
B
x
A
" 0x
Phase diagram'x
x
x
" 0x " 0x
0x Tx
Consider points A and B.
' ' , B A B A x x x x
'( ' ) '( ' ) ( "( ') 0)B A
rf x rf x f x
'( ) '( ' ) '( ) '( ' ) 0 B B A Ag x rf x g x rf x
Hence any point lying above the " 0x
curve has the property that " 0 'x x
The solution to the differential equation must
be either single-peaked, monotonic, or
single-troughed in ( )x t , '( )x t changes sign at
most once.
There is only 1 path going from0
x to Tx and takes exactly T .
8/3/2019 N20122 Mathematical Economics II 1 141
47/141
47
Example
1 2' { ( ) [ ( ), ] } 0, 0, [0, ] 0, [ ( ),0] ,
' [ ( ), ] ( ) '(.) 0, (0) 0
k s f k A f k r nk A A A A f k
A f k n f f
:k capital-labor ratios : saving rate : debt per capita
A : aid per capita
: retirement of principal
r: interest rate
n : labor growth rate
' 0 curve
' [ ( ), ] ( ) 0 A f k n
1 2
1 2
11 2
' 0 2
'( ) ( ) 0'( ) ( ) 0
'( )0 0, 0
A f k dk A d n d A f k dk A n d
A f k dA A
dk A n
' 0
I J
k
0 0k
, I J J I k k
'
'
[ ( ), ] ( )
[ ( ), ] ( ) 0
J J J J
I I I I
A f k n
A f k n
8/3/2019 N20122 Mathematical Economics II 1 141
48/141
48
' 0 curvek
1 2
1 2
11
' 0 2
' { ( ) [ ( ), ] } 0
{ '( ) [ '( ) ] } 0
{ [ '( ) '( )] } [ ( ) ] 0
[ '( ) '( )]0 iff '( )(1 ) 0
( )k
k s f k A f k r nk
s f k dk A f k dk A d rd d ndk
s f k A f k n dk s A r d
s f k A f k ndsf k A n
dk s r A
I
J
' 0k
k
' 0
' 0k
k
When k is small, '( )f k is large.
, I J I J
k k
'
'
{ ( ) [ ( ), ] }
{ ( ) [ ( ), ] } 0
J J J J J J J
I I I I I I I
k s f k A f k r nk
s f k A f k r nk k
8/3/2019 N20122 Mathematical Economics II 1 141
49/141
49
Stability
Take a Taylor series expansion around ( , )S Sk .
' ' ' '
' '( , ) ( ) ( ) ( ) ( )
' ' ' '' '( , ) ( ) ( ) ( ) ( )
S S S S
S S S S
S S S S
S S S S
S S S S S Sk k k k k k k k
S S S S S Sk k k k k k k k
k k k k
k k k k k k k k k
k k k k k k k
1 1 1
'{ '( ) '( )} '( )(1 ) 0
S
S
S S Sk k
ks f k A f k n sf k A n a
k
2 1
'( ) 0
S
S
k k
ks A r b
1 2' '( ) 0S
S
k k
A f k ak
2 2
'( )
8/3/2019 N20122 Mathematical Economics II 1 141
50/141
50
Calculus of Variation
Motivation
Find the shortest distance in the plane between the points, ( , ) and ( , )a A b B .
ds dx
21
2 2 2 22[( ) ( ) ] ( ) [1 ] 1 '( )dx
ds dt dx dt x t dt dt
dt1
22min [1 '( ) ] s.t. ( ) , ( )
b
a
x t dt x a A x b B
General problem1
0
0 0 1 1( ) [ , ( ), '( )] s.t. ( ) , ( )
t
x tt
optimize F t x t x t dt x t x x t x
Necessary condition for an optimum (Eulers Equation): '[ , *( ), *'( )]
[ , *( ), *'( )] xxdF t x t x t
F t x t x t dt
Example1
22
( )min [1 '( ) ] s.t. ( ) , ( )
b
x ta
x t dt x a A x b B 1
2 2
12 2
' 1
2 2
[ , ( ), '( )] [1 '( ) ]
1 '0 and [1 '( ) ] [2 '( )]2
[1 ' ]
x x
F t x t x t x t
F xF F x t x t x
x
(Eulers Equation): '[ , *( ), * '( )]
[ , *( ), * '( )] xx
dF t x t x t F t x t x t
dt
1
2 122 2 2 2 2 2 2 22
1
2 2
2 22 2 2 2
2 2
'
[1 ( ') ] '0 ' [1 ( ') ] ( ') [1 ( ') ] ( ') ( ')
[1 ( ') ]
(1 )( ') ( ') '( ) ( ) , are constants1 1
xd
x xc x c x x c x x c c x
dtx
c c dxc c x x x t k k x t kt h k h
c c dt
( ) 1,
( ) 1
x a ka h A a k A A B aB Abk h
x b kb h B b h B a b a b
8/3/2019 N20122 Mathematical Economics II 1 141
51/141
51
Optimal control theory
Problem:1
0
( )[ , ( ), ( )] (1)
t
u tt
Max f t x t u t dt
s.t. '( ) [ , ( ), ( )] (2)x t g t x t u t dynamic equation/ state equation/ transition equation
0 1
0 0
1
, fixed
( ) fixed (3)
( ) free
t t
x t x
x t
[ , ( ), ( )]f t x t u t
( ) :x t state variable
( ) :u t control variable (piecewise continuous)
,f g continuously differentiable
0t 1t
Let ( )t be any continuously differentiable function on 0 1[ , ]t t .
For any ( ), ( )x t u t satisfying (2) and (3), we have
1 1
0 0
1
0
[ , ( ), ( )] [ , ( ), ( )] ( )[ ( , ( ), ( )) '( )]
[ , ( ), ( )] ( ) [ , ( ), ( )] ( ) '( ) (4)
t t
t t
t
t
f t x t u t dt f t x t u t t g t x t u t x t dt
f t x t u t t g t x t u t t x t dt
Integrating by part the last term of (4),
1 1 1
1
0
0 0 0
1
0
1 1 0 0
1 1 0 0
( ) '( ) ( ) ( ) ( ) '( ) ( ) ( ) ( ) ( ) ( ) '( )
( ) ( ) ( ) ( ) ( ) '( )
t t tt
t
t t t
t
t
t x t dt t x t x t t dt t x t t x t x t t dt
t x t t x t x t t dt
(5)
1
0
1
0
1 1 0 0
(5) (4) [ , ( ), ( )]
[ , ( ), ( )] ( ) [ , ( ), ( )] ( ) '( ) ( ) ( ) ( ) ( ) (6)
t
t
t
t
f t x t u t dt
f t x t u t t g t x t u t x t t dt t x t t x t
Let *( )u t be the optimal control which maximizes the integral1
0
[ , ( ), ( )]
t
t
f t x t u t dt .
Let ( ) *( ) ( )u t u t ah t where ( ) is an arbitrary "fixed" function and is a parameterh t a .
Clearly, 0 ( ) *( )a u t u t .
Find a path
which maximize
the area/integral.
8/3/2019 N20122 Mathematical Economics II 1 141
52/141
52
Let 0 1( , ),y t a t t t denotes the state variable generated by (2) and (3) with control ( )u t .
This ( , ) y t a satisfies the following conditions: 0 0( , ) and ( ,0) *( )y t a x y t x t .
Let
1
0
( ) [ , ( , ), *( ) ( )]
t
t
J a f t y t a u t ah t dt
Using (6), 1
0
( ) [ , ( , ), *( ) ( )] ( ) [ , ( , ), *( ) ( )] ( , ) '( )
t
t
J a f t y t a u t ah t t g t y t a u t ah t y t a t dt
1 1 0 0( ) ( , ) ( ) ( , )t y t a t y t a
Since *( )u t is the optimal control, ( )J a assumes its maximum at 0 '(0) 0a J .
1
0
11 0
'( ) [ , ( , ), *( ) ( )] ( , ) [ , ( , ), *( ) ( )] ( )
( ) [ , ( , ), *( ) ( )] ( , ) ( ) [ , ( , ), *( ) ( )] ( )
(( , )( , ) '( ) ( ) ( )
t
y a u
t
y a u
a
J a f t y t a u t ah t y t a f t y t a u t ah t h t
t g t y t a u t ah t y t a t g t y t a u t ah t h t
dy tdy t a y t a t dt t t
da
0 , )a
da
As 00 0
( , )( , ) (initial condition) 0
dy t ay t a x
da
1
0
1 1
'( ) [ , ( , ), *( ) ( )] ( , ) [ , ( , ), *( ) ( )] ( )
( ) [ , ( , ), *( ) ( )] ( , ) ( ) [ , ( , ), *( ) ( )] ( )
( , ) '( ) ( ) ( , )
t
y a u
t
y a u
a a
J a f t y t a u t ah t y t a f t y t a u t ah t h t
t g t y t a u t ah t y t a t g t y t a u t ah t h t
y t a t dt t y t a
1
0
1 1
'(0) [ , ( ,0), *( )] ( ,0) [ , ( ,0), *( )] ( ) ( ) [ , ( ,0), *( )] ( ,0)
( ) [ , ( ,0), *( )] ( ) ( ,0) '( ) ( ) ( ,0)
= [ , ( ), *( )
t
y a u y a
t
u a a
x
J f t y t u t y t f t y t u t h t t g t y t u t y t
t g t y t u t h t y t t dt t y t
f t x t u t
1
0
1 1
] ( ) [ , ( ), *( )] '( ) ( ,0)
[ , ( ), *( )] ( ) [ , ( ), *( )] ( ) ( ) ( ,0) 0
t
x a
t
u u a
t g t x t u t t y t
f t x t u t t g t x t u t h t dt t y t
Since ( , 0), ( )a y t h t can be of any form, therefore the necessary conditions for '(0) 0J are
i) ' 0x xf g
ii) 0u uf g
iii) 1( ) 0t
8/3/2019 N20122 Mathematical Economics II 1 141
53/141
53
Sum up:
If the functions *( ), *( )u t x t maximizes (1) subject to (2) and (3), then there is a continuously
differentiable ( )t such that
i) state equation 0 0'( ) ( , ( ), ( )), ( )x t g t x t u t x t x
ii) co-state equation 1'( ) [ , ( ), ( )] ( ) [ , ( ), ( )] , ( ) 0x xt f t x t u t t g t x t u t t
iii) optimality condition 0 1[ , ( ), ( )] ( ) [ , ( ), ( )] 0 foru uf t x t u t t g t x t u t t t t
Short-cut formulas:
Problem:1
0( )
0 0
( , , )
. . ' ( , , )
( )
t
u tt
Optimize f t x u dt
s t x g t x u
x t x
Set up a Hamiltonian function: [ , , , ] ( , , ) ( , , )H t x u f t x u g t x u
Necessary conditions:
i) 0 [ 0]u u uH
H f gu
ii) ' [ ' ]x xH f gx
iii) 1( ) 0t transversality condition
iv) ' ( , , )H
H x g t x u
8/3/2019 N20122 Mathematical Economics II 1 141
54/141
54
Example1
2
0
max ( ) . . ' 1 , (0) 1, (1) freeu
x u dt s t x u x x
2( , , , ) (1 )H t x u x u u
Necessary conditions:
2
1 2 0 (1)
1 ' (2)
' 1 (3)
(1) 0
uH u
H u x
H
x
(4)
(3): '( ) 1 ( )t t t c Using (4) ( ) 1 (5)t t
2
12
1 1 1(1) : ( ) ( )
2 2(1 ) 2(1 )
1 1(2) : '( ) 1 1 ( )
4(1 ) 4(1 )
u t u t t t
x t u x t t ct t
1 1
1 5 1 5Using (0) 1 1 0 ( )
4(1 0) 4 4(1 ) 4 x c c x t t
t
8/3/2019 N20122 Mathematical Economics II 1 141
55/141
55
Example10
2
0
max 5 s.t. ' , (0) 2, (10) freeu
x dt x x u x x
2( , , , ) 5 ( )H t x u x x u
Necessary conditions:
2
2 0 (1)
' (2)
' (5 ) (3)
(10) 0
uH u
H x u x
H
x
(4)
(1)
10 10 10 10
0
5(3) : ' 5 ( ) 5
1
Using (10) 0 0 (10) 5 5 5 ( ) 5 5
(1) ( ) 0
(2) ' ( )
Using (0) 2 2 (0) ( ) 2
t t
t
t
t
t ce ce
ce ce c e t e
u t
x x x t ke
x x ke k x t e
10
2
0
max ( 5 ) . . ' , (0) 2, (10) freeu
u x dt s t x x u x x
2( , , , ) 5 ( )H t x u u x x u
Necessary conditions:
2
1 2 0 (1)
' (2)
' (5 ) (3)
(10) 0 (4)
u
HH u
u
H x H x u
H
x
10 10 10 10
5(3) ( ) 5
1
(4) 0 (10) 5 5 ( ) 5 (5 ) ( ) 5 5 (5)
t t
t t
t ce ce
ce c e t e e t e
10 10
1 1 1(5) (1) ( )
2 ( ) 2( 5 5 ) 10 10t tu t
t e e
' ( ) PtQ
y Py Q y t ceP
8/3/2019 N20122 Mathematical Economics II 1 141
56/141
56
2 2
10 10
1 1(2) ' '
10 10 10 10t t
x x x xe e
2 21 1
10 10
10
10
1 1( ) ( )
10 10 10 10
1( )
10 10 10
dt dt t t
t t
t
t
x t e e dt k x t e e dt k
e e
e x t e k
e
Using (0) 2x , we have10 10
0
10 0 10
1 12 (0) 2
10 1010 10 10 10
e e x e k k
e e
10 10
10 10
1 1( ) 2
10 10 10 10 10 10
t
t
e e x t e
e e
8/3/2019 N20122 Mathematical Economics II 1 141
57/141
57
5
2 2
1
max ( )
. . ' , (1) 2
uux u x dt
s t x x u x
2 2
( ) H ux u x x u
Necessary conditions:
2 0 (1)
' ( 2 ) 2
u
X
H x u
H u x u x
(2)
(5) 0 (3)
' x x u
(4)
(1) 2 (5)x
(1) 2u x (6)
(4) ' (4')u x x
(4') (6) 2( ' ) 2 ' 3 x x x x x (7)
Differentiate (7) with respect to ' 2 " 3 't x x (8)
(6) (2) ' 2 (2 ) 3 3u x u x x u (9)
(8), (4) and (9) 2 " 3 ' 3 3( ' ) 2 " 3 ' 6 3 ' 2 " 6 0 " 3 0x x x x x x x x x x x x x
Characteristic equation:2 3 0 3r r
3 3
1 2( ) t t x t c e c e (10)
Differentiating (10) with respect to 3 31 2' 3 3t t
t x c e c e
(4): 3 3 3 31 2 1 2
' ( 3 3 ) ( )t t t t u x x c e c e c e c e 3 3
1 2( 3 1) ( 3 1)t t
u c e c e (11)
(6) and (11)3 3 3 3
1 2 1 22 2 ( 3 1) ( 3 1)t t t t u x c e c e c e c e
3 3 3 3
1 2 1 2( ) (2 3 2 1) (2 3 2 1) (2 3 3) (2 3 3)t t t t t c e c e c e c e
Using (3), we have 5 3 5 31 2(5) (2 3 3) (2 3 3) 0c e c e (12)
8/3/2019 N20122 Mathematical Economics II 1 141
58/141
58
(5) and (10) 3 31 2
(1) 2 x c e c e (13)
(12) and (13)
5 3 5 3
1
3 32
(2 3 3) (2 3 3) 0
2
ce e
ce e
5 3
3 5 3 5 3
1 4 3 4 3 4 3 4 35 3 5 3
3 3
5 3
3 5 3 5 3
2 4 3 4 3 4 3 4 35 3 5 3
3 3
0 (2 3 3)2 2(2 3 3) 2
(2 3 3) (2 3 3) (7 4 3)(2 3 3) (2 3 3)
(2 3 3) 0
2 2(2 3 3) 2
(2 3 3) (2 3 3) (7 4 3)(2 3 3) (2 3 3)
ee e e
ce e e ee e
e e
e
e e ec
e e e ee e
e e
Solution:3 3
1 2( )t t
x t c e c e
3 3
1 2( ) ( 3 1) ( 3 1)t tu t c e c e
3 3
1 2( ) (2 3 3) (2 3 3)t tt c e c e
8/3/2019 N20122 Mathematical Economics II 1 141
59/141
59
Example: Find the shortest distance between a point ( , )At A and the line bt t .
A x ds dx
dt
t
A Bt t
Problem:
2
2
min min (1 ( ') s.t. ( )
min (1 )
s.t. '
( )
( ) free
B B
A A
B
A
t t
A A
t t
t
ut
A A
B
ds x dt x t X
u dt
x u
x t X
x t
Hamiltonian:21 H u u
Necessary conditions:
20 (1)
(1 )
' (2)
' 0 (3)
( ) 0 (4)
u
X
B
uH
u
x u
H
t
(3) ( )t c (5)
(4) and (5) ( ) 0t
Substitute2
( ) 0 (1) 0 01
ut u
u
0 (2) ' 0 ( )u x x t k ( )x t is a horizontal line.
2 2 2 2 2( ) ( ) [1 ( ) ]( ) 1 ( ')dx
ds dt dx dt x dt dt
8/3/2019 N20122 Mathematical Economics II 1 141
60/141
60
Example: Optimal price path
( )P t : price @ t
( )Q t : cumulative sales @ t
( )C Q : unit production cost '( ) 0C Q
( ) ( )f P g Q : the rate that a new product can be sold1
'( ) 0; "( ) 0
'( ) 0 for
f P f P
g Q Q Q
Problem:
T
( )0
max [ ( )] ( ) ( )P t
P C Q f P g Q dt
s.t. ' ( ) ( )Q f P g Q
0(0) 0Q Q
( , , , ) ( , , , ) [ ( )] ( ) ( ) ( ) ( ) ( ) ( )[ ( ) ]H t x u H t Q P P C Q f P g Q f P g Q f P g Q P C Q
Optimal conditions:
( ) ( ) [ ( ) ] '( ) 0PH g Q f P P C Q f P (1)
' ( ) [ ( ) ] '( ) ( ) '( )
( ) ( ) '( ) [ ( ) ] '( )
QH f P P C Q g Q g Q C Q
f P g Q C Q P C Q g Q
(2)
( ) 0T (3)
' ( ) ( )Q f P g Q (4)
(1): ( )( ) [ ( ) ] '( ) 0 ( )'( )
f Pf P P C Q f P P C Qf P
( )( ) ( )
'( )
f Pt C Q P
f P (5)
Differentiate (5) with respect to t:
2
2
2 2
'( ) '( ) '( ) ( ) "( ) '( )'( ) '( ) '( ) '( )
'( )
' '( ) ( ) "( ) ' ( ) "( )' '( ) ' ' '( ) ' ' ''( ) '( )
f P f P P t f P f P P t t C Q Q t P t
f P
P f P f P f P P f P f PC Q Q P C Q Q P P f P f P
2
( ) "( )' '( ) ' ' 2
'( )
f P f PC Q Q P
f P
(6)
: control variable
: state variable
P
Q
8/3/2019 N20122 Mathematical Economics II 1 141
61/141
61
(5) (2) ' ( ) ( ) '( ) [ ( ) ] '( )
( )( ) ( ) '( ) ( ) ( ) '( )
'( )
( )( ) ( ) '( ) '( )'( )
f P g Q C Q P C Q g Q
f Pf P g Q C Q P C Q C Q P g Q
f P
f P
f P g Q C Q g Qf P
(2')
(2) and (6)
2
( ) "( ) ( )'( ) ' ' 2 ( ) ( ) '( ) '( )
'( )'( )
f P f P f PC Q Q P f P g Q C Q g Q
f Pf P
2
2
2
( ) "( ) ( )' ( ) ( ) '( ) ( ) ( ) ' 2 ( ) ( ) '( ) '( )
'( )'( )
( ) ( ) (+)
( ) "( ) '( )' 2 ( )
'( )'( )
f P f P f PQ f P g Q C Q f P g Q P f P g Q C Q g Q
f Pf P
f P f P g QP f P
f Pf P
( )
'( ) '( )sign P t sign g Q
When the market is expanding (i.e. 1Q Q ), ( )P t is increasing.
When the market is shrinking (i.e. 1Q Q ), ( )P t is decreasing.
8/3/2019 N20122 Mathematical Economics II 1 141
62/141
62
4 Interpretation of ( )t
1
0
0 0
0 0
Define ( , ) max ( , , )
s.t. '( ) ( , , )
( )
t
t
V x t f t x u dt
x t g t x u
x t x
(1)
Let *( )u t be the optimal control and *( )x t be the optimal path of ( )x t .
1 1
0 0
1
0
0 0( , ) ( , *, *) ( , *, *) [ ( , *, *) ']
= ( , *, *) ( , *, *) '] (2)
t t
t t
t
t
V x t f t x u dt f t x u g t x u x dt
f t x u g t x u x dt
where ( )t is the corresponding continuously differentiable multiplier function
Integrate the last term of (2) by parts:
1
0
1
0
0 0
1 1 0 0
( , ) ( , *, *) ( , *, *) ']
( , *, *) ( , *, *) * '] ( ) * ( ) ( ) ( ) (3)
t
t
t
t
V x t f t x u g t x u x dt
f t x u g t x u x dt t x t t x t
Now, using the same ( )t , define
1
0
0 0
0 0
( , ) max ( , , )
s.t. '( ) ( , , )
( )
t
t
V x a t f t x u dt
x t g t x u
x t x a
1 1 1
1
0
0 0 0
1 1 0 0
Integrate by parts:
Let ( ) ( ) '( ) , ' ( ) ' ( )
' ( ) ( ) ' ( ) ( ) ( ) ( ) '
t t tt
t
t t t
udv uv vdu duv udv vdu uv duv udv vdu
u t t du t dt dv x dt v t dv x dt x t
x dt t x t x dt t x t t x t x dt
0 0*( ) ( ) x t x t
8/3/2019 N20122 Mathematical Economics II 1 141
63/141
63
1 1
0 0
1
0
0 0
1 1 0 0
( , ) ( , , ) ( , , ) ( , , ) ']
( , , ) ( , , ) '] ( ) ( ) ( ) ( ) (4)
t t
t t
t
t
V x a t f t x u dt f t x u g t x u x dt
f t x u g t x u x dt t x t t x t a
where ( ) and ( ) are optimal for the new problem. x t u t
1
0
1
0
0 0 0 0 1 1 0 0
1 1 0 0
0 1 1
( , ) ( , ) ( , , ) ( , , ) '] ( ) ( ) ( ) ( )
( , *, *) ( , *, *) * '] ( ) *( ) ( ) ( )
[ ' * * ' *] ( ) ( )[ ( )
t
t
t
t
V x a t V x t f t x u g t x u x dt t x t t x t a
f t x u g t x u x dt t x t t x t
f g x f g x dt t a t x t
1
0
1*( )] (5)
t
tx t
Expand ( , , ) ( , , ) around ( *, *)f t x u g t x u x u
( , , ) ( , , ) ( , *, *) ( , *, *) ( , *, *) ( , *, *) ( *)
* * ( *) ...
x x
u u
f t x u g t x u f t x u g t x u f t x u g t x u x x
f g u u
1
0
0 0 0 0
0 1 1 1
(5) ( , ) ( , ) [ * * ']( *) [ * *]( *)
( ) ( )[ ( ) *( )] (6)
t
x x u u
t
V x a t V x t f g x x f g u u dt
t a t x t x t
Note that
1
' ( * *)
*, *, and satisfy * * 0
( ) 0
x x
u u
f g
x u f g
t
0 0 0 0 0
0 0 0 00
0 0 0 0 0 00 0 0
00
(6) ( , ) ( , ) ( )
( , ) ( , )( )
( , ) ( , ) ( , )( ) lim ( , )
xa
V x a t V x t t a
V x a t V x t t
aV x a t V x t V x t
t V x t a x
8/3/2019 N20122 Mathematical Economics II 1 141
64/141
64
Principle of optimality: If *( ) and *( )u t x t are optimal paths, then any sub-paths are also optimal.
*( )x t
#x
t#
0 1t t t
Suppose we follow#*( ), * ( ) tox t u t t and stop. We then try to solve the following problem:
1
#
# #max ( , , ) s.t. ' ( , , ) and ( )
t
t
f t x u dt x g t x u x t x
Let ( )x t be the solution to this problem.
Claim; ( ) *( ) x t x t
Suppose not.
That is1
#
# #( ) maximizes ( , , ) s.t. ' ( , , ) and ( )
t
t
x t f t x u dt x g t x u x t x so that
1 1
# #
( , , ) ( , *, *)
t t
t t
f t x u dt f t x u dt
Now define# #
0 0
# #
1 1
*( ) for [ , ] *( ) for [ , ]( ) and ( )
( ) for ( , ] ( ) for ( , ]
x t t t t u t t t t x t u t
x t t t t u t t t t
Clearly
# #1 1 1
# #0 0 0
( , , ) ( , *, *) ( , , ) ( , *, *) ( , *, *)
t t tt t
t t tt t
f t x u dt f t x u dt f t x u dt f t x u dt f t x u dt
This implies that *( )x t cannot be the optimal path of the original problem. Contradiction.
Note that
#1 1
#0 0
( , , ) ( , , ) ( , , )t tt
t t t
f t x u dt f t x u dt f t x u dt
In general, # # #[ ( ), ] ( )xV
V x t t t x
8/3/2019 N20122 Mathematical Economics II 1 141
65/141
65
Example:
x : productive capital [which decays at a constant proportionate rate 0b ]( )P x : profit rate that can be earned with the stock of productive capital '(0) 0 and " 0P P
u : gross investment rate
( )C x : cost of investment '(0) 0 and " 0C C
Problem:
0( )
0
max [ ( ) ( )] s.t. ' , (0) 0, 0
T
rt
u te P x C u dt x u bx x x u
[Note that as 0u and 0 0 ( ) 0 [0, ] x x t t T ].
( , , , ) [ ( ) ( )] ( )rtH t x u e P x C u u bx
Necessary conditions:
'( ) 0 (1)
' '( ) (2)
( ) 0 (3)
'
rt
urt
x
H e C u
H e P x b
T
x u bx
(4)
(1) '( ) ( )rtC u e t
(2): ' '( ) ' '( )rt rt e P x b b e P x
( ) ( )
( ) ( )
( ) ( ) (
( ' ) '( ) [ '( ) ( )] '( ( ))
( ) '( ( )) ( ) ( ) '( ( ))
( ) '( ( )) ( )
T T
bt r b t bs r b s
t t
T TT
bs r b s bT bt r b s
tt t
T
bt r b s r b t bt r
t
e b e P x e s b s ds e P x s ds
e s e P x s ds e T e t e P x s ds
e t e P x s ds e e t e
) ( )
( )( )
'( ( ))
( ) '( ( ))
T
b t r b s
t
T
rt r b s t
t
e P x s ds
e t e P x s ds
Finally, we have ( )( )'( ( )) '( ( ))T
r b s t
t
C u t e P x s ds
marginal cost
marginal cost of investment current value of the marginal value of capital
Value @ t of a marginal unit of capital the discount stream of future marginal profits it generates
The calculation reflects the fact that capital decays, and therefore at each time s t a unit of capitalcontributes only a fraction bse of what it contributed originally @ t.
8/3/2019 N20122 Mathematical Economics II 1 141
66/141
66
Sufficiency
Problem:1
0
0 0
max ( , , )
. . ' ( , , ), ( )
t
ut
f t x u dt
s t x g t x u x t x
Theorem: Suppose that
i) ( , , )f t x u and ( , , )g t x u are both differentiable concave functions of ( , )x u ,
ii) *( ), *( ), ( ) x t u t t satisfy the necessary conditions,
iii) ( ) and ( ) x t t are continuous with ( ) 0t for all t if ( , , )g t x u is nonlinear in orx u , or both,
then *( ), *( ) x t u t solve the problem.
Note:
i) If the function ( , , )g t x u is linear in ( , )x u , then ( )t may assume any sign.
ii) If ( , , )f t x u is concave while ( , , )g t x u is convex and ( ) 0t , then the necessary conditionswill also be sufficient for optimality.
8/3/2019 N20122 Mathematical Economics II 1 141
67/141
67
6 Fixed endpoint problem
1
0
( )
0 0
1 1
max ( , , )
. . ' ( , , )( )
( )
t
u tt
f t x u dt
s t x g t x u x t x
x t x
We need to modify the equations on p. 46-47 in the notes.
1
0
( ) [ , ( , ), *( ) ( )] ( ) [ , ( , ), *( ) ( ) ( , ) '( )
t
t
J a f t y t a u t ah t t g t y t a u t ah t y t a t dt
1 1 0 0( ) ( , ) ( ) ( , )t y t a t y t a
1
0
11 0
'( ) [ , ( , ), *( ) ( )] ( , ) [ , ( , ), *( ) ( )] ( )
( ) [ , ( , ), *( ) ( )] ( , ) ( ) [ , ( , ), *( ) ( )] ( )
(( , )( , ) '( ) ( ) ( )
t
y a u
t
y a u
a
J a f t y t a u t ah t y t a f t y t a u t ah t h t
t g t y t a u t ah t y t a t g t y t a u t ah t h t
dy tdy t a y t a t dt t t
da
0, )a
da
As 00 0
( , )( , ) (initial condition) 0
dy t ay t a x
da
Also1
1 1
( , )
( , ) (endpoint condition) 0
dy t a
y t a x da
1
0
'(0) = [ , ( ), *( )] ( ) [ , ( ), *( )] '( ) ( ,0)
[ , ( ), *( )] ( ) [ , ( ), *( )] ( ) 0
t
x x a
t
u u
J f t x t u t t g t x t u t t y t
f t x t u t t g t x t u t h t dt
Since ( , 0), ( )a y t h t can be of any form, therefore the necessary conditions for '(0) 0J are
i) ' 0x xf g
ii) 0u uf g
No transversality condition.
8/3/2019 N20122 Mathematical Economics II 1 141
68/141
68
Production planning problem
Objective: produce B units of goods in a period of T in a way such that the sum of production
cost and storage cost is minimized.
Problem:2
1 2( )
0
min ( ) s.t. '( ) , (0) 0, ( ) , 0
T
u tc u c x dt x t u x x T B u
2
1 2( , , , )H t x u c u c x u
Necessary conditions:
1
2
0 2 0 (1)
' (2)
' (3)
u
x
H c u
H c
H x u
(2) 2 2 1'( ) ( )t c t c t k
2 1 2 1
1 1 1 1
( )( )(1) ( )
2 2 2 2
c t k c t k tu t
c c c c
22 1 2 12
1 1 1 1
(3) '( ) ( )2 2 4 2
c k c k x t u t x t t t k
c c c c
Using the initial condition and endpoint conditions,
22 12 2 2
1 1
(0) 0 (0) (0) 04 2
c k x k k k
c c
2 2
22 1 2 1 1 2 2 12 1
1 1 1 1 1 1
2( ) ( ) ( )
4 2 4 2 2 4 2
c k c T k T k T c T c T Bc x T B T T k B B k
c c c c c c T
2 12 1 2
2( )
2
c T Bct c t k c t
T
2 1
2 1 2 2 2 2
1 1 1 1 1 1 1
2
(2 )2( )2 2 2 2 2 4 4
c T Bc
c t k c t c t c T c t T B BTu tc c c c c c T c T
2 1
2 2 22 1 2 2 2 2
2
1 1 1 1 1 1 1
2
( )2( ) (0)
4 2 4 2 4 4 4
c T Bc
c k c c c T c t t T T B Bt x t t t k t t t t t
c c c c c c T c T
Note that we assume( ) 0, [0, ]u t t T
8/3/2019 N20122 Mathematical Economics II 1 141
69/141
69
Example:
Two factors, capital ( )K t and extractive resource ( )R t are used to produce a good Q according to the
production function1 a aQ AK R
, where 0 1a .
The output can be consumed, yielding utility ( ) lnU C C , or it may be invested.
Problem:
( ), ( )0
0
1
0
max ln
. . ' , (0) , ( ) 0
' , (0) , ( ) 0
0, 0
T
C t R t
a a
C dt
s t x R x x x T
K AK R C K K K T
C R
2 state variables: ,x K
2 control variables: ,C R
Define( )
( )( )
R ty t
K t resource-capital ratio [ ( ) ( ) ( )]R t y t K t
Problem:
( ), y( )0
0
0
max ln
. . ' , (0) , ( ) 0
' , (0) , ( ) 00, 0
T
C t t
a
C dt
s t x Ky x x x T
K AKy C K K K T C y
1 2ln ( )a H C Ky AKy C
Necessary conditions:
2
1
1 2
1
2 1 2
10 (1)
0 (2)
' 0 (3)'
C
a
y
x
a
K
HC
H K a AKy
H H y Ay
(4)
' (5)
' (6)a
x KY
K AKy C
Since 0 ( ) 0 MU C t t
Since marginal product ofR as 0 0 ( ) 0 R R y t t
8/3/2019 N20122 Mathematical Economics II 1 141
70/141
70
(2): 11 2 ( 0) (2')a
a Ay K
1
1 12 1
1
1 1 1
2 1
(7)
[(1 ) '] ' (1 ) '' [ ' 0 by (3)]
a
a
a a a
y
aAy aA
a y y y a y y
aA aA
Using (7)
1
2
1
12
(1 ) '
' (1 ) '(8)
a
a
a y y
a yaA
y y
aA
1 22 2 22
'(2 ') (4) ' (1 ) (9)a a a a aa Ay y Ay aAy Ay a Ay
1 1
1
' '(8) and (9) (1 ) (1 )
1 1 (10)
a
a a
a
a
a a
y y dya a Ay A A dt
y y y
dy y A dt At k aAt ak y y a y aAt ak
1
( ) 1( )
( )
aR ty t
K t aAt ak
Also, the capital-output ratio
( ) 1is linear in , i.e. it is growing linearly at the rate
( ) a aK t K aAt ak ak
at t aQ t AKy Ay A A
2
2 1
' 1 (1 )(10) (9) (1 ) (1 )a aa Ay aak at k
atA
2
2 2
22 1 2
2 1
1
2 1
1 1
1 1
(1 ) (1 )ln ( ) ln( )
( ) ( )
1(1) ( ) ( ) ( )
( )
ak