ugarchfit - Weighted Ljung-Box Test and ARCH LM Test

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

ugarchfit - Weighted Ljung-Box Test and ARCH LM Test

Lukas Halbeisen
Hi all,

I am currently working on my master thesis and have to evaluate some
GARCH-Modells.

What I am wondering:

How are the lags picked up for the Weighted LB-Test as well as for the
ARCH LM Test.

For instance I have fitted a GARCH(3,3)-Modell:

Weighted Ljung-Box Test on Standardized Squared Residuals
------------------------------------
                          statistic  p-value
Lag[1]                       9.052 0.002624
Lag[2*(p+q)+(p+q)-1][17]    14.593 0.070672
Lag[4*(p+q)+(p+q)-1][29]    20.134 0.129824
d.o.f=6


The lags are calculated by the formular 2*(p+q)+(p+q)-1. But what is the
reason? Is there any reference where this kind of calculation has been
described?


Weighted ARCH LM Tests

------------------------------------
              Statistic Shape Scale P-Value
ARCH Lag[7]      1.375 0.500 2.000  0.2409
ARCH Lag[9]      1.462 1.485 1.796  0.6476
ARCH Lag[11]     2.596 2.440 1.677  0.6695


Same for the ARCH-LM modell. I have 6 d.o.f. but why does rugarch
consider the lags 7,9 and 11 as the most important ones?

Thank you and best regards,
Lukas

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.
Reply | Threaded
Open this post in threaded view
|

Re: ugarchfit - Weighted Ljung-Box Test and ARCH LM Test

prof.amit.mittal
Try and get hold of basic econometric texts like today and enders (two
separate authors) it is a well described methodology

On Sun 21 Oct, 2018, 12:55 AM Lukas Halbeisen, <[hidden email]> wrote:

> Hi all,
>
> I am currently working on my master thesis and have to evaluate some
> GARCH-Modells.
>
> What I am wondering:
>
> How are the lags picked up for the Weighted LB-Test as well as for the
> ARCH LM Test.
>
> For instance I have fitted a GARCH(3,3)-Modell:
>
> Weighted Ljung-Box Test on Standardized Squared Residuals
> ------------------------------------
>                           statistic  p-value
> Lag[1]                       9.052 0.002624
> Lag[2*(p+q)+(p+q)-1][17]    14.593 0.070672
> Lag[4*(p+q)+(p+q)-1][29]    20.134 0.129824
> d.o.f=6
>
>
> The lags are calculated by the formular 2*(p+q)+(p+q)-1. But what is the
> reason? Is there any reference where this kind of calculation has been
> described?
>
>
> Weighted ARCH LM Tests
>
> ------------------------------------
>               Statistic Shape Scale P-Value
> ARCH Lag[7]      1.375 0.500 2.000  0.2409
> ARCH Lag[9]      1.462 1.485 1.796  0.6476
> ARCH Lag[11]     2.596 2.440 1.677  0.6695
>
>
> Same for the ARCH-LM modell. I have 6 d.o.f. but why does rugarch
> consider the lags 7,9 and 11 as the most important ones?
>
> Thank you and best regards,
> Lukas
>
> _______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions
> should go.
>
--

______________________________

Amit Mittal
Pursuing Ph.D. in Finance and Accounting
Indian Institute of Management, Lucknow
Visit my SSRN author page:
http://ssrn.com/author=2665511
* Top 10% Downloaded Author on SSRN
Mob: +91 7525023664

This message has been sent from a mobile device. I may contact you again.

_________________

        [[alternative HTML version deleted]]

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.
Reply | Threaded
Open this post in threaded view
|

Re: ugarchfit - Weighted Ljung-Box Test and ARCH LM Test

Lukas Halbeisen
Hi Amit,

thanks for your quick reply. :)

I invested today another 4 hours to find some adequate information,
which lags should be tested. As well, I checked "Applied Econometric
Time Series" from Enders to find some appropiate information to this
matter, but with no result.

Would you be so kind to give me some more details to the references you
are talking about? Or can you tell me how the methodology is called,
that would be very helpful for me.

Thank you very much and best regards,
Lukas

Am 20.10.2018 um 21:38 schrieb Amit Mittal:

> Try and get hold of basic econometric texts like today and enders (two
> separate authors) it is a well described methodology
>
> On Sun 21 Oct, 2018, 12:55 AM Lukas Halbeisen, <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     Hi all,
>
>     I am currently working on my master thesis and have to evaluate some
>     GARCH-Modells.
>
>     What I am wondering:
>
>     How are the lags picked up for the Weighted LB-Test as well as for
>     the
>     ARCH LM Test.
>
>     For instance I have fitted a GARCH(3,3)-Modell:
>
>     Weighted Ljung-Box Test on Standardized Squared Residuals
>     ------------------------------------
>                               statistic  p-value
>     Lag[1]                       9.052 0.002624
>     Lag[2*(p+q)+(p+q)-1][17]    14.593 0.070672
>     Lag[4*(p+q)+(p+q)-1][29]    20.134 0.129824
>     d.o.f=6
>
>
>     The lags are calculated by the formular 2*(p+q)+(p+q)-1. But what
>     is the
>     reason? Is there any reference where this kind of calculation has
>     been
>     described?
>
>
>     Weighted ARCH LM Tests
>
>     ------------------------------------
>                   Statistic Shape Scale P-Value
>     ARCH Lag[7]      1.375 0.500 2.000  0.2409
>     ARCH Lag[9]      1.462 1.485 1.796  0.6476
>     ARCH Lag[11]     2.596 2.440 1.677  0.6695
>
>
>     Same for the ARCH-LM modell. I have 6 d.o.f. but why does rugarch
>     consider the lags 7,9 and 11 as the most important ones?
>
>     Thank you and best regards,
>     Lukas
>
>     _______________________________________________
>     [hidden email] <mailto:[hidden email]>
>     mailing list
>     https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>     -- Subscriber-posting only. If you want to post, subscribe first.
>     -- Also note that this is not the r-help list where general R
>     questions should go.
>
> --
>
> ______________________________
>
> Amit Mittal
> Pursuing Ph.D. in Finance and Accounting
> Indian Institute of Management, Lucknow
> Visit my SSRN author page:
> http://ssrn.com/author=2665511
> * Top 10% Downloaded Author on SSRN
> Mob: +91 7525023664
>
> This message has been sent from a mobile device. I may contact you again.
>
> _________________
>

        [[alternative HTML version deleted]]

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.
Reply | Threaded
Open this post in threaded view
|

Re: ugarchfit - Weighted Ljung-Box Test and ARCH LM Test

Josh Segal
Lukas,

The rugarch vignette references this paper, have you read it?
http://cecas.clemson.edu/~cgallag/publications/FishGal.pdf

For more details you can follow references therein.

Josh

On Mon, Oct 22, 2018 at 2:29 PM Lukas Halbeisen <[hidden email]>
wrote:

> Hi Amit,
>
> thanks for your quick reply. :)
>
> I invested today another 4 hours to find some adequate information,
> which lags should be tested. As well, I checked "Applied Econometric
> Time Series" from Enders to find some appropiate information to this
> matter, but with no result.
>
> Would you be so kind to give me some more details to the references you
> are talking about? Or can you tell me how the methodology is called,
> that would be very helpful for me.
>
> Thank you very much and best regards,
> Lukas
>
> Am 20.10.2018 um 21:38 schrieb Amit Mittal:
> > Try and get hold of basic econometric texts like today and enders (two
> > separate authors) it is a well described methodology
> >
> > On Sun 21 Oct, 2018, 12:55 AM Lukas Halbeisen, <[hidden email]
> > <mailto:[hidden email]>> wrote:
> >
> >     Hi all,
> >
> >     I am currently working on my master thesis and have to evaluate some
> >     GARCH-Modells.
> >
> >     What I am wondering:
> >
> >     How are the lags picked up for the Weighted LB-Test as well as for
> >     the
> >     ARCH LM Test.
> >
> >     For instance I have fitted a GARCH(3,3)-Modell:
> >
> >     Weighted Ljung-Box Test on Standardized Squared Residuals
> >     ------------------------------------
> >                               statistic  p-value
> >     Lag[1]                       9.052 0.002624
> >     Lag[2*(p+q)+(p+q)-1][17]    14.593 0.070672
> >     Lag[4*(p+q)+(p+q)-1][29]    20.134 0.129824
> >     d.o.f=6
> >
> >
> >     The lags are calculated by the formular 2*(p+q)+(p+q)-1. But what
> >     is the
> >     reason? Is there any reference where this kind of calculation has
> >     been
> >     described?
> >
> >
> >     Weighted ARCH LM Tests
> >
> >     ------------------------------------
> >                   Statistic Shape Scale P-Value
> >     ARCH Lag[7]      1.375 0.500 2.000  0.2409
> >     ARCH Lag[9]      1.462 1.485 1.796  0.6476
> >     ARCH Lag[11]     2.596 2.440 1.677  0.6695
> >
> >
> >     Same for the ARCH-LM modell. I have 6 d.o.f. but why does rugarch
> >     consider the lags 7,9 and 11 as the most important ones?
> >
> >     Thank you and best regards,
> >     Lukas
> >
> >     _______________________________________________
> >     [hidden email] <mailto:[hidden email]>
> >     mailing list
> >     https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> >     -- Subscriber-posting only. If you want to post, subscribe first.
> >     -- Also note that this is not the r-help list where general R
> >     questions should go.
> >
> > --
> >
> > ______________________________
> >
> > Amit Mittal
> > Pursuing Ph.D. in Finance and Accounting
> > Indian Institute of Management, Lucknow
> > Visit my SSRN author page:
> > http://ssrn.com/author=2665511
> > * Top 10% Downloaded Author on SSRN
> > Mob: +91 7525023664
> >
> > This message has been sent from a mobile device. I may contact you again.
> >
> > _________________
> >
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions
> should go.
>

        [[alternative HTML version deleted]]

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.
Reply | Threaded
Open this post in threaded view
|

Re: ugarchfit - Weighted Ljung-Box Test and ARCH LM Test

prof.amit.mittal
In reply to this post by Lukas Halbeisen
Ruey tsay - analysis of financial time series
Enders - Applied Economet time series
Greene

------------
Amit Mittal
PhD in Finance and Accounting (tbd)
IIM Lucknow
http://ssrn.com/author=2665511
*Top 10%, downloaded author since July 2017
------------
Sent from my Outlook for Android
https://aka.ms/ghei36

________________________________
From: Lukas Halbeisen <[hidden email]>
Sent: Monday, October 22, 2018 11:59:01 PM
To: [hidden email]
Cc: [hidden email]
Subject: Re: [R-SIG-Finance] ugarchfit - Weighted Ljung-Box Test and ARCH LM Test


Hi Amit,

thanks for your quick reply. :)

I invested today another 4 hours to find some adequate information, which lags should be tested. As well, I checked "Applied Econometric Time Series" from Enders to find some appropiate information to this matter, but with no result.

Would you be so kind to give me some more details to the references you are talking about? Or can you tell me how the methodology is called, that would be very helpful for me.

Thank you very much and best regards,
Lukas


Am 20.10.2018 um 21:38 schrieb Amit Mittal:
Try and get hold of basic econometric texts like today and enders (two separate authors) it is a well described methodology

On Sun 21 Oct, 2018, 12:55 AM Lukas Halbeisen, <[hidden email]<mailto:[hidden email]>> wrote:
Hi all,

I am currently working on my master thesis and have to evaluate some
GARCH-Modells.

What I am wondering:

How are the lags picked up for the Weighted LB-Test as well as for the
ARCH LM Test.

For instance I have fitted a GARCH(3,3)-Modell:

Weighted Ljung-Box Test on Standardized Squared Residuals
------------------------------------
                          statistic  p-value
Lag[1]                       9.052 0.002624
Lag[2*(p+q)+(p+q)-1][17]    14.593 0.070672
Lag[4*(p+q)+(p+q)-1][29]    20.134 0.129824
d.o.f=6


The lags are calculated by the formular 2*(p+q)+(p+q)-1. But what is the
reason? Is there any reference where this kind of calculation has been
described?


Weighted ARCH LM Tests

------------------------------------
              Statistic Shape Scale P-Value
ARCH Lag[7]      1.375 0.500 2.000  0.2409
ARCH Lag[9]      1.462 1.485 1.796  0.6476
ARCH Lag[11]     2.596 2.440 1.677  0.6695


Same for the ARCH-LM modell. I have 6 d.o.f. but why does rugarch
consider the lags 7,9 and 11 as the most important ones?

Thank you and best regards,
Lukas

_______________________________________________
[hidden email]<mailto:[hidden email]> mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.
--

______________________________

Amit Mittal
Pursuing Ph.D. in Finance and Accounting
Indian Institute of Management, Lucknow
Visit my SSRN author page:
http://ssrn.com/author=2665511
* Top 10% Downloaded Author on SSRN
Mob: +91 7525023664

This message has been sent from a mobile device. I may contact you again.

_________________

        [[alternative HTML version deleted]]

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.
Reply | Threaded
Open this post in threaded view
|

Re: ugarchfit - Weighted Ljung-Box Test and ARCH LM Test

Lukas Halbeisen
In reply to this post by Josh Segal
Josh,

I read it at least twice and there is not a single sentence about which
lags are optimal to test against Ljung-Box-Test or Li-Mak-Test. Or do I
miss anything?

Am 22.10.2018 um 20:34 schrieb Josh Segal:

> Lukas,
>
> The rugarch vignette references this paper, have you read it?
> http://cecas.clemson.edu/~cgallag/publications/FishGal.pdf
>
> For more details you can follow references therein.
>
> Josh
>
> On Mon, Oct 22, 2018 at 2:29 PM Lukas Halbeisen <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     Hi Amit,
>
>     thanks for your quick reply. :)
>
>     I invested today another 4 hours to find some adequate information,
>     which lags should be tested. As well, I checked "Applied Econometric
>     Time Series" from Enders to find some appropiate information to this
>     matter, but with no result.
>
>     Would you be so kind to give me some more details to the
>     references you
>     are talking about? Or can you tell me how the methodology is called,
>     that would be very helpful for me.
>
>     Thank you very much and best regards,
>     Lukas
>
>     Am 20.10.2018 um 21:38 schrieb Amit Mittal:
>     > Try and get hold of basic econometric texts like today and
>     enders (two
>     > separate authors) it is a well described methodology
>     >
>     > On Sun 21 Oct, 2018, 12:55 AM Lukas Halbeisen,
>     <[hidden email] <mailto:[hidden email]>
>     > <mailto:[hidden email] <mailto:[hidden email]>>> wrote:
>     >
>     >     Hi all,
>     >
>     >     I am currently working on my master thesis and have to
>     evaluate some
>     >     GARCH-Modells.
>     >
>     >     What I am wondering:
>     >
>     >     How are the lags picked up for the Weighted LB-Test as well
>     as for
>     >     the
>     >     ARCH LM Test.
>     >
>     >     For instance I have fitted a GARCH(3,3)-Modell:
>     >
>     >     Weighted Ljung-Box Test on Standardized Squared Residuals
>     >     ------------------------------------
>     >                               statistic  p-value
>     >     Lag[1]                       9.052 0.002624
>     >     Lag[2*(p+q)+(p+q)-1][17]    14.593 0.070672
>     >     Lag[4*(p+q)+(p+q)-1][29]    20.134 0.129824
>     >     d.o.f=6
>     >
>     >
>     >     The lags are calculated by the formular 2*(p+q)+(p+q)-1. But
>     what
>     >     is the
>     >     reason? Is there any reference where this kind of
>     calculation has
>     >     been
>     >     described?
>     >
>     >
>     >     Weighted ARCH LM Tests
>     >
>     >     ------------------------------------
>     >                   Statistic Shape Scale P-Value
>     >     ARCH Lag[7]      1.375 0.500 2.000  0.2409
>     >     ARCH Lag[9]      1.462 1.485 1.796  0.6476
>     >     ARCH Lag[11]     2.596 2.440 1.677  0.6695
>     >
>     >
>     >     Same for the ARCH-LM modell. I have 6 d.o.f. but why does
>     rugarch
>     >     consider the lags 7,9 and 11 as the most important ones?
>     >
>     >     Thank you and best regards,
>     >     Lukas
>     >
>     >     _______________________________________________
>     > [hidden email] <mailto:[hidden email]>
>     <mailto:[hidden email]
>     <mailto:[hidden email]>>
>     >     mailing list
>     > https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>     >     -- Subscriber-posting only. If you want to post, subscribe
>     first.
>     >     -- Also note that this is not the r-help list where general R
>     >     questions should go.
>     >
>     > --
>     >
>     > ______________________________
>     >
>     > Amit Mittal
>     > Pursuing Ph.D. in Finance and Accounting
>     > Indian Institute of Management, Lucknow
>     > Visit my SSRN author page:
>     > http://ssrn.com/author=2665511
>     > * Top 10% Downloaded Author on SSRN
>     > Mob: +91 7525023664
>     >
>     > This message has been sent from a mobile device. I may contact
>     you again.
>     >
>     > _________________
>     >
>
>             [[alternative HTML version deleted]]
>
>     _______________________________________________
>     [hidden email] <mailto:[hidden email]>
>     mailing list
>     https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>     -- Subscriber-posting only. If you want to post, subscribe first.
>     -- Also note that this is not the r-help list where general R
>     questions should go.
>

        [[alternative HTML version deleted]]

_______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.