Partial Adjustment Model

The partial adjustment model comprises two parts, a static part to describe how the
desired amount is determined and the dynamic partial adjustment process:
) (
1
*
1
1 0
*

=
+ + =
t t t t
t t t
y y y y
u x y

Where y* is the desired level of y !y substitutin" the e#pression for y* into the other
e$uation we obtain the followin" estimatin" e$uation:
t t t t
u x y y + + + =
1 1 0
) 1 (
We can estimate this e$uation as a "eneral %&'( model as follows:
t t t t t
v x x y y + + + + =
1 ) * 1 1 0

+n this case the followin" restriction would be imposed if partial adjustment occurred:
0
)
=
+n addition we could "et estimates of the parameters in the ori"inal e$uation
containin" the desired level of y, as well as the adjustment parameter , +n the above
case:



0 0
1 *
1 1
) 1 ( ) 1 (
=
=
= =
The adjustment parameter , measures the speed of adjustment and lies between 0 and
1 The closer it is to 1 the faster the speed of adjustment %n e#ample of this model is
the (intner 'ividend-%djustment .odel
The Error Correction Model
The error correction model (/0.) is an e#ample of a short-run dynamic model,
which is used to model both economic and financial time series +t is e#pressed in first
differences (
1
=
t t t
y y y
), e#cept the error correction term +t is based on the
%&'( model a"ain and can be derived from this model, with the addition of a
specific restriction
t t t t t
v x x y y + + + + =
1 ) * 1 1 0

To produce the /0., firstly you have to subtract y
t-1
from both sides of the %&'(
e$uation:
t t t t t t t t
v y x x y y y y + + + + = =
1 1 ) * 1 1 0 1

The ne#t step is to e#press the x in difference form This involves addin" and
subtractin"
1 * t
x
from the ri"ht hand side of the above e$uation
t t t t t t t t
v x y x x x y y + + + + + =
1 * 1 1 ) 1 * * 1 1 0

0ollectin" terms "ives:
t t t t t
v x y x y + + + + + =
1 ) * 1 1 * 0
) ( ) 1 (
+n order to produce the /0., we need to assume that the coefficient on y
t-1
is e$ual to
minus the coefficient on x
t-1
That means:
1
) ( 1
) * 1
) * 1
= + +
+ =


The sum of the coefficients, e#cludin" the constant, must sum to one in the %&'(
model, for the /0. to apply The /0. is usually written with 1 as the coefficient on
the error correction term, to "ive the followin":
) ( ) 1 (
) (
) * 1
1 1 * 0


+ = =
+ + =
t t t t t
u x y x y
The above /0. is the short-run relationship between y and x The lon"-run
relationship can be formed in the usual way, althou"h instead of assumin" all
differenced terms e$ual 0, we assume that they "row at a constant rate g This "ives:
*
) 1 (
*
) 1 ( *) * (
*) * (
* 0
* 0
* 0
x
g
y
g x y
x y g g
+
+
=
+ =
+ =

+f the ori"inal model was:
t t
kx y * * = which in lo"s is
t t
x k y * lo" lo" * lo" + =
Then in terms of the above lon"-run e#pression, when antilo""ed:
2
) 1 (
e#p3
* 0

g
k
+
=
The term k can be interpreted as the lon"-run relationship between y and x ie if y is
consumption and x is wealth, k would be the avera"e propensity to consume from
wealth