Page 1 of 1

Lee-Strazicich (LM) Unit Root Test

Posted: Wed Sep 03, 2008 11:08 pm
by prakash
Hello sir, I need your help for understanding the output result of Lee and Strazicich unit root test. I am getting my test results in the below given format. basically, I not able to understand which is LM coefficient in the output table below and then how to calculate lembdas from this result. I will be greatful to you for your help and cooperation.


Lee-Strazicich Unit Root Test, Series SDP
Regression Run From 1986:01 to 2004:01
Observations 13
Crash Model with 2 breaks
With 0 chosen from 5

Variable Coefficient T-Stat
S{1} -0.6210 -1.8787
Constant -17.6988 -0.8250
D(1989:01) -23.9401 -3.0851
D(1997:01) 61.9471 7.2655

Lee-Strazicich Unit Root Test, Series SDP
Regression Run From 1986:01 to 2004:01
Observations 13
Trend Break Model with 2 breaks
With 5 chosen from 5

Variable Coefficient T-Stat
S{1} -2.7279 -14.2298
Constant 9.1735 5.5809
D(1989:01) -133.4288 -0.0000
DT(1989:01) 134.5588 0.0000
D(1990:01) 9.2571 2.2997
DT(1990:01) -157.3393 -0.0000

Re: Lee-Strazicich (LM) Unit Root Test

Posted: Thu Sep 04, 2008 9:51 am
by TomDoan
prakash wrote:Hello sir, I need your help for understanding the output result of Lee and Strazicich unit root test. I am getting my test results in the below given format. basically, I not able to understand which is LM coefficient in the output table below and then how to calculate lembdas from this result. I will be greatful to you for your help and cooperation.

Code: Select all

Lee-Strazicich Unit Root Test, Series SDP
Regression Run From 1986:01 to 2004:01
Observations          13
Crash Model with 2 breaks
With 0 chosen from 5

Variable     Coefficient               T-Stat
S{1}                  -0.6210             -1.8787
Constant        -17.6988             -0.8250
D(1989:01)    -23.9401            -3.0851
D(1997:01)     61.9471             7.2655
The test statistic is the T-Stat on S{1}, in this case the -1.8787.

Lee-Strazicich (LM) Unit Root Test

Posted: Fri Sep 05, 2008 1:40 am
by prakash
Dear Tom, Thanks for your help, I have one more query in this regard and that is in my result whether the the t- statistics which I have got are LM(tuo) or LM (rho). if you could put some light on it I will be thakful to you

Thanks