Chapter 21
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
Time Series Models
There are no exercises or applications in Chapter 21.
131
This
preview
has intentionally blurred sections.
Sign up to view the full version.
Chapter 22
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
Nonstationary Data
Exercise
1.
The autocorrelations are simple to obtain just by multiplying out v
t
2
, v
t
v
t-1
and so on. The
autocovariances are 1+
θ
1
2
+
θ
2
2
, -
θ
2
(1 -
θ
2
), -
θ
2
, 0, 0, 0... which provides the autocorrelations by division by
the first of these.
The partial autocorrelations are messy, and can be obtained by the Yule Walker
equations.
Alternatively (and much more simply), we can make use of the observation in Section 21.2.3
that the partial autocorrelations for the MA(2) process mirror tha autocorrelations for an AR(2).
Thus, the
results in Section 21.2.3 for the AR(2) can be used directly.
Applications
1.
ADF Test
+-----------------------------------------------------------------------+
| Ordinary
least squares regression
Weighting variable = none
|
| Dep. var. = R
Mean=
8.212678571
, S.D.=
.7762719558
|
| Model size: Observations =
56, Parameters =
6, Deg.Fr.=
50 |
| Residuals:
Sum of squares= .9651001703
, Std.Dev.=
.13893 |
| Fit:
R-squared=
.970881, Adjusted R-squared =
.96797 |
| Model test: F[
5,
50] =
333.41,
Prob value =
.00000 |
| Diagnostic: Log-L =
34.2439, Restricted(b=0) Log-L =
-64.7739 |
|
LogAmemiyaPrCrt.=
-3.846, Akaike Info. Crt.=
-1.009 |

This is the end of the preview.
Sign up
to
access the rest of the document.
- Spring '10
- Dr.Fang
- Least Squares, Regression Analysis, Standard Error, Akaike Info, Error |t-ratio |P, |Variable | Coefficient
-
Click to edit the document details