Page 1 of 1

PDL

Posted: Thu Feb 17, 2011 10:11 am
by RDS
Below is the typical output of a PDL equation.
Is there a command to store the sum of lags and its standard errors?

Thnak you very much. Roberto

Dependent Variable: DXX
Method: Least Squares
Date: 02/16/11 Time: 22:35
Sample (adjusted): 1985Q1 2010Q4
Included observations: 104 after adjustments
Newey-West HAC Standard Errors & Covariance (lag truncation=4)

Variable Coefficient Std. Error t-Statistic Prob.

C 0.767509 0.166929 4.597833 0.0000
PDL01 0.224442 0.046465 4.830367 0.0000
PDL02 -0.064022 0.043347 -1.476945 0.1430
PDL03 -0.027902 0.062641 -0.445433 0.6570
PDL04 -0.016538 0.043524 -0.379978 0.7048
PDL05 -0.039767 0.045472 -0.874540 0.3840
PDL06 0.030217 0.085821 0.352092 0.7255
PDL07 0.016712 0.014000 1.193736 0.2356
PDL08 -0.003799 0.020167 -0.188356 0.8510

R-squared 0.813380 Mean dependent var 2.657135
Adjusted R-squared 0.797664 S.D. dependent var 0.996112
S.E. of regression 0.448069 Akaike info criterion 1.314823
Sum squared resid 19.07273 Schwarz criterion 1.543664
Log likelihood -59.37078 F-statistic 51.75684
Durbin-Watson stat 0.318701 Prob(F-statistic) 0.000000

Lag Distribution of DX(-12) i Coefficient Std. Error t-Statistic

. *| 0 0.26056 0.05331 4.88743
. * | 1 0.22444 0.04646 4.83037
. * | 2 0.13252 0.05367 2.46916

Sum of Lags 0.61752 0.06725 9.18187

Lag Distribution of EX_LIQN40US(-12) i Coefficient Std. Error t-Statistic

*. | 0 -0.01061 0.05342 -0.19861
. * | 1 0.03294 0.04996 0.65920
* . | 2 -0.01654 0.04352 -0.37998
*. | 3 -0.01317 0.04253 -0.30978
. *| 4 0.09771 0.03539 2.76118

Sum of Lags 0.09033 0.02930 3.08256

Re: PDL

Posted: Thu Feb 17, 2011 10:14 am
by EViews Gareth
The only thing you can do is freeze the stats view of the equation and then retrieve those numbers from the resultant table. Something like:

Code: Select all

freeze(mytable) myeq.stats scalar sum = mytable(32,4)
where 32,4 is the table cell containing the sum.

Re: PDL

Posted: Thu Feb 17, 2011 10:43 am
by RDS
Unfortunately, it is not one equation. It is a loop of several equations and all of them have different locations.

Best regards,

Roberto

Re: PDL

Posted: Thu Feb 17, 2011 10:47 am
by EViews Gareth
You'll have to figure out a way to calculate its position.

Re: PDL

Posted: Fri Feb 18, 2011 1:23 am
by RDS
In the future, it would be great if views creates a command which allows to extract the sum of lags and its coefficients. Best regards, Roberto

PDL

Posted: Fri Feb 18, 2011 8:08 am
by EViews Gareth
You can get the pdl coefficients with eq.@coefs, it is the standard errors that aren't available