Economy

USING TIME SERIES MODELS

On  the class web site, you will find two files:

• An Excel file called y1y2y3.XLS
• An EXCEL file called www.XLS

PART ONE:

Open the file called y1y2y3.XLS

There are three series, labelled Y1, Y2 and Y3.

• One of these is an AR(1) model:                           Yt = 0.7Yt-1 + εt 
• Another is an AR(2) model:         :                   Yt = 0.7Yt-1   - 0.49Yt-2 + εt 
• The third is an ARMA(1,1) model:                  Yt =  - 0.7Yt-1   + εt  - 0.7  εt-1

The objective of this exercise is for you to use autocorrelation coefficients and partial autocorrelation coefficients to identify these series, and so to decide which of Y1, Y2 and Y3 is an AR(1), which an AR(2), and which is an ARMA(1,1).

To do this, open up Eviews, and do the tests using the built-in graphical output. Activate the view choice on the menu in eviews, followed by the wfile page on the dialog. Then mark Correlogram.

Answer- Part One

For y1 series, the correlogram as follows:

[pic 1][pic 2][pic 3]

Since ACF is declining and PACF has 1 spike, this means that AR(1) is appropriate for this.

For y2 series, the correlogram as follows:

[pic 4][pic 5][pic 6]

Since the both ACF and PACF are declining, this means that ARMA(1) is appropriate for this.

For y3 series, the correlogram as follows:

[pic 7][pic 8][pic 9]

Since ACF is declining and PACH has 2 spikes, AR(2) is appropriate for this.

PART TWO

The file on disk is an EXCEL spreadsheet file. It contains data on one variable, WWW.