0% found this document useful (0 votes)
11 views

Autocorrelation Graphs and Tables

The document discusses various methods to address autocorrelation in time series regression models, including: 1) Autocorrelation graphs and correlation matrices to examine the correlation between errors over time. 2) Durbin-Watson, Durbin h, and Breusch-Godfrey tests to formally test for autocorrelation. 3) Prais-Winsten and Cochrane-Orcutt transformations to correct for first-order autocorrelation in the regression errors.

Uploaded by

cosgapacyril21
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
11 views

Autocorrelation Graphs and Tables

The document discusses various methods to address autocorrelation in time series regression models, including: 1) Autocorrelation graphs and correlation matrices to examine the correlation between errors over time. 2) Durbin-Watson, Durbin h, and Breusch-Godfrey tests to formally test for autocorrelation. 3) Prais-Winsten and Cochrane-Orcutt transformations to correct for first-order autocorrelation in the regression errors.

Uploaded by

cosgapacyril21
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

AUTOCORRELATION GRAPHS AND TABLES

Original Regression

. reg elc gdppc mob


Source | SS df MS Number of obs = 24
-------------+------------------------------ F( 2, 21) = 1216.77
Model | 613387.439 2 306693.72 Prob > F = 0.0000
Residual | 5293.18084 21 252.056231 R-squared = 0.9914
-------------+------------------------------ Adj R-squared = 0.9906
Total | 618680.62 23 26899.1574 Root MSE = 15.876
------------------------------------------------------------------------------
elc | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gdppc | .5076964 .0374269 13.57 0.000 .429863 .5855298
mob | 1.001007 .3506034 2.86 0.009 .2718869 1.730127
_cons | 105.6931 20.8132 5.08 0.000 62.40966 148.9765
------------------------------------------------------------------------------

. predict error, resid

. plot error year

30.2386 +
| *
|
| *
|
| *
R | * *
e | * *
s |
i | * *
d | * * * *
u | *
a | *
l | * * *
s | * *
| * * *
|
|
|
|
-34.4452 + *
+----------------------------------------------------------------+
1991 year 2014
Correlation Matrix

. pwcorr error error_t_1 error_t_2 error_t_3 error_t_4 error_t_5 error_t_6, sig

| error error_~1 error_~2 error_~3 error_~4 error_~5 error_~6


-------------+---------------------------------------------------------------
error | 1.0000
|
|
error_t_1 | 0.7083 1.0000
| 0.0002
|
error_t_2 | 0.1954 0.7083 1.0000
| 0.3834 0.0002
|
error_t_3 | -0.2011 0.1954 0.6584 1.0000
| 0.3820 0.3834 0.0009
|
error_t_4 | -0.2608 -0.2011 0.1207 0.6336 1.0000
| 0.2668 0.3820 0.6023 0.0020
|
error_t_5 | -0.2032 -0.2608 -0.1296 0.2728 0.7356 1.0000
| 0.4042 0.2668 0.5861 0.2446 0.0002
|
error_t_6 | -0.3564 -0.2032 -0.0694 0.1996 0.5201 0.7438 1.0000
| 0.1466 0.4042 0.7778 0.4127 0.0225 0.0003
|

Durbin- Watson Test


. tsset year
time variable: year, 1991 to 2014
delta: 1 unit

. dwstat
Durbin-Watson d-statistic( 3, 24) = .5480991
Durbin M Test

. reg error error_t_1 gdppc mob


Source | SS df MS Number of obs = 23
-------------+------------------------------ F( 3, 19) = 8.50
Model | 2881.7712 3 960.5904 Prob > F = 0.0009
Residual | 2147.2261 19 113.0119 R-squared = 0.5730
-------------+------------------------------ Adj R-squared = 0.5056
Total | 5028.9973 22 228.590786 Root MSE = 10.631
------------------------------------------------------------------------------
error | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
error_t_1 | .8191529 .1635065 5.01 0.000 .4769298 1.161376
gdppc | -.0412529 .0273026 -1.51 0.147 -.0983978 .0158921
mob | .4399056 .2538263 1.73 0.099 -.0913589 .9711701
_cons | 22.80843 15.37617 1.48 0.154 -9.374269 54.99113
------------------------------------------------------------------------------

Breusch and Godfrey Test

. reg error error_t_1 gdppc mob

Source | SS df MS Number of obs = 23


-------------+------------------------------ F( 3, 19) = 8.50
Model | 2881.7712 3 960.5904 Prob > F = 0.0009
Residual | 2147.2261 19 113.0119 R-squared = 0.5730
-------------+------------------------------ Adj R-squared = 0.5056
Total | 5028.9973 22 228.590786 Root MSE = 10.631
------------------------------------------------------------------------------
error | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
error_t_1 | .8191529 .1635065 5.01 0.000 .4769298 1.161376
gdppc | -.0412529 .0273026 -1.51 0.147 -.0983978 .0158921
mob | .4399056 .2538263 1.73 0.099 -.0913589 .9711701
_cons | 22.80843 15.37617 1.48 0.154 -9.374269 54.99113
------------------------------------------------------------------------------
Prais- Winsten Transformation
. prais elc gdppc mob
Prais-Winsten AR(1) regression -- iterated estimates
Source | SS df MS Number of obs = 24
-------------+------------------------------ F( 2, 21) = 175.63
Model | 40685.5868 2 20342.7934 Prob > F = 0.0000
Residual | 2432.37683 21 115.827468 R-squared = 0.9436
-------------+------------------------------ Adj R-squared = 0.9382
Total | 43117.9636 23 1874.69407 Root MSE = 10.762
------------------------------------------------------------------------------
elc | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gdppc | .5888332 .0563663 10.45 0.000 .471613 .7060534
mob | .3630296 .5083172 0.71 0.483 -.6940739 1.420133
_cons | 61.55362 33.45643 1.84 0.080 -8.022846 131.1301
-------------+----------------------------------------------------------------
rho | .807919
------------------------------------------------------------------------------
Durbin-Watson statistic (original) 0.548099
Durbin-Watson statistic (transformed) 1.564997

Cochrane- Orcutt Method


. prais elc gdppc mob, corc
Cochrane-Orcutt AR(1) regression -- iterated estimates
Source | SS df MS Number of obs = 23
-------------+------------------------------ F( 2, 20) = 167.88
Model | 40547.0959 2 20273.5479 Prob > F = 0.0000
Residual | 2415.30214 20 120.765107 R-squared = 0.9438
-------------+------------------------------ Adj R-squared = 0.9382
Total | 42962.398 22 1952.83627 Root MSE = 10.989
------------------------------------------------------------------------------
elc | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gdppc | .5697767 .0728295 7.82 0.000 .417857 .7216963
mob | .4656674 .5664566 0.82 0.421 -.7159405 1.647275
_cons | 76.43407 49.6628 1.54 0.139 -27.16073 180.0289
-------------+----------------------------------------------------------------
rho | .797268
------------------------------------------------------------------------------
Durbin-Watson statistic (original) 0.548099
Durbin-Watson statistic (transformed) 1.500850

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy