Anda di halaman 1dari 5

ARIMA Forecasting

Graeme Walsh
May 14, 2013
Abstract
The following document is meant to give an brief indication of how to
go about producing point and interval forecasts from an ARIMA model.
The examples in the document are based on an AR(1) model. The same
method presented here can be used produce forecasts from other ARIMA
models. Note, however, that the equations will be different for each model
and that a little bit of algebra is required. This document is a far from
complete treatment of the subject of forecasting ARIMA models. I refer
the interested reader to Pankratz (1983)1 for more details.

Introduction
Suppose we have identified an AR(1) model to represent a time-series. The
model can be written as:
zt = C + 1 zt1 + at

(1)

where C is a constant and at is an error term.


Further suppose that we estimate the model and find that 1 = 0.690 and
C = 1.92091.
Assume also that the available data consists of 60 observations and that
z60 = 6.2.

Point Forecasts
The easiest way to generate point forecasts is to work with the model in difference form as shown in Equation 1.
We need to write an expression for our first forecast. We do this by altering
the subscripts in Equation 1 as follows:
zt+1 = C + 1 zt + at+1

(2)

1 Pankratz, A., 1983. Forecasting with univariate Box-Jenkins models: Concepts and cases.
New York: Wiley.

Now, our forecast of zt+1 , denoted zt (1), is the conditional mathematical expectation of zt+1 given information at time t, or, It . That is:
zt (1)

= E(zt+1 |It )

(3)

= C + 1 z t

(4)

Notice that at+1 is unknown at time t, hence we set its expected value equal to
zero.
A similar process can be used to find more point forecasts. Lets find the
next three point forecasts.
The first one is given by:
zt+2

C + 1 zt+1 + at+2

(5)

zt (2)

= C + 1 zt (1) + at+2

(6)

= E(zt+1 |It )

(7)

= C + 1 zt (1)

(8)

zt+3

= C + 1 zt+2 + at+3

(9)

= C + 1 zt (2) + at+3

(10)

zt (3)

= E(zt+2 |It )

(11)

= C + 1 zt (2)

(12)

= C + 1 zt+3 + at+4

(13)

= C + 1 zt (3) + at+4

(14)

= E(zt+3 |It )

(15)

= C + 1 zt (3)

(16)

The second one is given by:

The third one is given by:


zt+4
zt (4)

Example
Using the data, we can now create four point forecasts. They are found as
follows.
From Equation 4:
z61 = z60 (1)

= C + 1 z60

(17)

1.92091 + (0.690)(6.2)

(18)

6.198910

(19)

From Equation 8:
z62 = z60 (2)

= C + 1 z61

(20)

1.92091 + (0.690)(6.198910)

(21)

6.1982

(22)

From Equation 12:


z63 = z60 (3)

= C + 1 z62

(23)

1.92091 + (0.690)(6.1982)

(24)

6.1977

(25)

= C + 1 z63

(26)

1.92091 + (0.690)(6.1977)

(27)

6.1973

(28)

From Equation 16:


z64 = z60 (4)

Random Shock Form


To make interval forecasts we no longer work in difference form. Instead, we
write the model in random shock form. That is, we write the model in its MA
representation.
Our AR(1) model can be written as an infinite MA series:
zt = + at + 1 at1 + 21 at2 + 31 at3 +

(29)

Now let 0 = 1, 1 = 1 , 2 = 21 , 3 = 31 and so on. That is, let the


weights equal the coefficients on all of the MA terms.

Example
Using our data, this means that:
0

(30)

1 = 0.690

21
31

(31)
2

= (0.690) = 0.4761

(32)

= (0.690)3 = 0.3285

(33)

Standard Deviation of the Forecast Errors


The standard deviation of the forecast errors is given by:
2
[et (l)] = a (1 + 12 + 22 + + +l1
)1/2

(34)

where a is the standard deviation of the shocks. In practice, the RMSE can
replace a .

Example
Assume that the RMSE for our estimated AR(1) is 3.37903. The estimated standard deviation of the forecast errors for the first four forecasts can be calculated
as follows.
The first one is:

[et (1)]

=
a (1)1/2

(35)

(36)

3.37903

The second one is:


=
a (1 + 12 )1/2

[et (2)]

(37)
2 1/2

3.37903(1 + (0.690) )

(38)

4.1053

(39)

The third one is:

[et (3)]

a (1 + 12 + 22 )1/2

(40)
2

2 1/2

3.37903(1 + (0.690) + (0.4761) )

(41)

4.4093

(42)

The fourth one is:

[et (4)]

=
a (1 + 12 + 22 + 32 ))1/2

(43)

3.37903(1 + (0.690)2 + (0.4761)2 + (0.3285)2 )1/2

(44)

4.5469

(45)

Forecast Confidence Intervals


Now lets construct forecast intervals around each of the point forecasts that we
found earlier. We can create 95% confidence intervals using:
zt (l) 1.96
[et (l)]

(46)

For 80% confidence intervals we use:


zt (l) 1.28
[et (l)]

(47)

Example
Using our data, the 80% confidence interval forecasts are as follows:
z60 (1)
6.198910

1.28
[e60 (1)]

(48)

1.28(3.37903)

(49)

This gives us the interval forecasts: (1.8738, 10.5241)

z60 (2)

1.28
[e60 (2)]

(50)

6.198910

1.28(4.1053)

(51)

This gives us the interval forecasts: (0.9441, 11.4537)

z60 (3)

1.28
[e60 (3)]

(52)

6.198910

1.28(4.4093)

(53)

This gives us the interval forecasts: (0.5550, 11.8428)

z60 (4)

1.28
[e60 (4)]

(54)

6.198910

1.28(4.5469)

(55)

This gives us the interval forecasts: (0.3789, 12.0189)

Anda mungkin juga menyukai