Estimating the T-S Garch model

[John C. Frain]

I am trying to estimate a T-S Garch model with the following code -

library(fGarch)
myFinCenter =  "GMT"
dframe=read.csv(file="loss.csv")
dframe[1:5,]
loss=as.timeSeries(dframe)
head(loss)
tail(loss)
fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0,
include.delta=FALSE,trace=FALSE)
summary(fit)
fit2 = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0,
include.delta=FALSE,leverage=FALSE)

Output is at the end of the email.  I can estimate a model with
nonzero leverage (gamma1 non-zero) but I have problems when I try to
force gamma1 to be zero.  (See message at end of output.  Is there a
problem or have I not understood something.
I am using Windows XP,  R 2.7.0 and fGarch 260.72.  The data-set
loss.csv is attached.


--
John C Frain
Trinity College Dublin
Dublin 2
Ireland
www.tcd.ie/Economics/staff/frainj/home.htm
mailto:frainj@...
mailto:frainj@...


############## OUTPUT ##############################

> library(fGarch)
> myFinCenter =  "GMT"
> dframe=read.csv(file="loss.csv")
> dframe[1:5,]
           X       loss
1 1988-01-05 -3.2854337
2 1988-01-06 -2.6706190
3 1988-01-07  0.3662351
4 1988-01-08 -1.8030495
5 1988-01-11  1.1893279
> loss=as.timeSeries(dframe)
> head(loss)
                 TS.1
1988-01-05 -3.2854337
1988-01-06 -2.6706190
1988-01-07  0.3662351
1988-01-08 -1.8030495
1988-01-11  1.1893279
1988-01-12 -0.1981119
> tail(loss)
                 TS.1
2008-01-24 -4.8668943
2008-01-25  0.7808780
2008-01-28  1.7767990
2008-01-29 -1.1739951
2008-01-30 -0.5321964
2008-01-31  1.3622478
> fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0, include.delta=FALSE,trace=FALSE)
> summary(fit)

Title:
 GARCH Modelling

Call:
 garchFit(formula = ~aparch(1, 1), data = loss@Data, delta = 1,
    include.delta = FALSE, trace = FALSE)

Mean and Variance Equation:
 ~arma(0, 0) + ~aparch(1, 1)

Conditional Distribution:
 dnorm

Coefficient(s):
        mu       omega      alpha1      gamma1       beta1
-0.0624601   0.0274496   0.0816745  -0.3396468   0.9100114

Error Analysis:
        Estimate  Std. Error  t value Pr(>|t|)
mu     -0.062460    0.011934   -5.234 1.66e-07 ***
omega   0.027450    0.005269    5.210 1.89e-07 ***
alpha1  0.081675    0.009555    8.548  < 2e-16 ***
gamma1 -0.339647    0.047851   -7.098 1.27e-12 ***
beta1   0.910011    0.011797   77.139  < 2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Log Likelihood:
 6630.546    normalized:  1.316368

Standadized Residuals Tests:
                                Statistic p-Value
 Jarque-Bera Test   R    Chi^2  5720.432  0
 Shapiro-Wilk Test  R    W      NA        NA
 Ljung-Box Test     R    Q(10)  102.3393  0
 Ljung-Box Test     R    Q(15)  108.0036  4.440892e-16
 Ljung-Box Test     R    Q(20)  118.3874  5.551115e-16
 Ljung-Box Test     R^2  Q(10)  80.30754  4.369838e-13
 Ljung-Box Test     R^2  Q(15)  86.34728  4.738099e-12
 Ljung-Box Test     R^2  Q(20)  88.63123  1.28495e-10
 LM Arch Test       R    TR^2   66.55288  1.405827e-09

Information Criterion Statistics:
      AIC       BIC       SIC      HQIC
-2.630751 -2.624274 -2.630753 -2.628482

Description:
 Thu May 01 20:58:17 2008 by user: John C Frain

> fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0, include.delta=FALSE,leverage=FALSE)

Series Initialization:
 ARMA model:                arma
 Formula mean:              ~ arma(0, 0)
 GARCH model:               aparch
 Formula var:               ~ aparch(1, 1)
 ARMA Order:                0 0
 Max ARMA Order:            0
 GARCH Order:               1 1
 Max GARCH Order:           1
 Maximum Order:             1
 h.start:                   2
 llh.start:                 1
 Length of Series:          5037
 Recursion Init:            mci
 Series Scale:              1.000351

Parameter Initialization:
 Initial Parameters:          $params
 Limits of Transformations:   $U, $V
 Which Parameters are Fixed?  $includes
 Parameter Matrix:
                       U           V      params includes
    mu     -4.830854e-01   0.4830854 -0.04830854     TRUE
    omega   1.000703e-06 100.0703024  0.10007030     TRUE
    alpha1  1.000000e-08   1.0000000  0.10000000     TRUE
    gamma1 -1.000000e+00   1.0000000  0.10000000    FALSE
    beta1   1.000000e-08   1.0000000  0.80000000     TRUE
    delta   0.000000e+00   2.0000000  1.00000000    FALSE
    skew    1.000000e-01  10.0000000  1.00000000    FALSE
    shape   1.000000e+00  20.0000000  4.00000000    FALSE
 Index List of Parameters to be Optimized:
    mu  omega alpha1  beta1
     1      2      3      5
 Persistence:                  0.8797885

Iteration Path:



Now NLMINB


Error in gamma[i] : object is not subsettable
>


_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.

[michal miklovic]

Hi,

it is a bug and it has already been solved in the development version of fGarch. Rmetrics development web is at: http://r-forge.r-project.org/projects/rmetrics
where you can find the latest versions of all packages. Download the zipped Windows binaries and install them in the following order:
fUtilities, fEcofin, fCalendar, fSeries, fImport, fBasics, fBonds, fArma, fGarch, fTrading, fExtremes, fNonlinear, fOptions, fAsianOptions, fExoticOptions, fUnitRoots, fMultivar, fCopulae, fRegression, fAssets, fPortfolio, Rmetrics
and the estimation should work fine.

Best regards,

Michal


----- Original Message ----
From: John Frain <frainj@...>
To: R-SIG-Finance mailing list <r-sig-finance@...>
Sent: Thursday, May 1, 2008 11:23:47 PM
Subject: [R-SIG-Finance] Estimating the T-S Garch model

I am trying to estimate a T-S Garch model with the following code -

library(fGarch)
myFinCenter =  "GMT"
dframe=read.csv(file="loss.csv")
dframe[1:5,]
loss=as.timeSeries(dframe)
head(loss)
tail(loss)
fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0,
include.delta=FALSE,trace=FALSE)
summary(fit)
fit2 = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0,
include.delta=FALSE,leverage=FALSE)

Output is at the end of the email.  I can estimate a model with
nonzero leverage (gamma1 non-zero) but I have problems when I try to
force gamma1 to be zero.  (See message at end of output.  Is there a
problem or have I not understood something.
I am using Windows XP,  R 2.7.0 and fGarch 260.72.  The data-set
loss.csv is attached.


--
John C Frain
Trinity College Dublin
Dublin 2
Ireland
www.tcd.ie/Economics/staff/frainj/home.htm
mailto:frainj@...
mailto:frainj@...


############## OUTPUT ##############################

> library(fGarch)
> myFinCenter =  "GMT"
> dframe=read.csv(file="loss.csv")
> dframe[1:5,]
           X       loss
1 1988-01-05 -3.2854337
2 1988-01-06 -2.6706190
3 1988-01-07  0.3662351
4 1988-01-08 -1.8030495
5 1988-01-11  1.1893279
> loss=as.timeSeries(dframe)
> head(loss)
                 TS.1
1988-01-05 -3.2854337
1988-01-06 -2.6706190
1988-01-07  0.3662351
1988-01-08 -1.8030495
1988-01-11  1.1893279
1988-01-12 -0.1981119
> tail(loss)
                 TS.1
2008-01-24 -4.8668943
2008-01-25  0.7808780
2008-01-28  1.7767990
2008-01-29 -1.1739951
2008-01-30 -0.5321964
2008-01-31  1.3622478
> fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0, include.delta=FALSE,trace=FALSE)
> summary(fit)

Title:
 GARCH Modelling

Call:
 garchFit(formula = ~aparch(1, 1), data = loss@Data, delta = 1,
    include.delta = FALSE, trace = FALSE)

Mean and Variance Equation:
 ~arma(0, 0) + ~aparch(1, 1)

Conditional Distribution:
 dnorm

Coefficient(s):
        mu       omega      alpha1      gamma1       beta1
-0.0624601   0.0274496   0.0816745  -0.3396468   0.9100114

Error Analysis:
        Estimate  Std. Error  t value Pr(>|t|)
mu     -0.062460    0.011934   -5.234 1.66e-07 ***
omega   0.027450    0.005269    5.210 1.89e-07 ***
alpha1  0.081675    0.009555    8.548  < 2e-16 ***
gamma1 -0.339647    0.047851   -7.098 1.27e-12 ***
beta1   0.910011    0.011797   77.139  < 2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Log Likelihood:
 6630.546    normalized:  1.316368

Standadized Residuals Tests:
                                Statistic p-Value
 Jarque-Bera Test   R    Chi^2  5720.432  0
 Shapiro-Wilk Test  R    W      NA        NA
 Ljung-Box Test     R    Q(10)  102.3393  0
 Ljung-Box Test     R    Q(15)  108.0036  4.440892e-16
 Ljung-Box Test     R    Q(20)  118.3874  5.551115e-16
 Ljung-Box Test     R^2  Q(10)  80.30754  4.369838e-13
 Ljung-Box Test     R^2  Q(15)  86.34728  4.738099e-12
 Ljung-Box Test     R^2  Q(20)  88.63123  1.28495e-10
 LM Arch Test       R    TR^2   66.55288  1.405827e-09

Information Criterion Statistics:
      AIC       BIC       SIC      HQIC
-2.630751 -2.624274 -2.630753 -2.628482

Description:
 Thu May 01 20:58:17 2008 by user: John C Frain

> fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0, include.delta=FALSE,leverage=FALSE)

Series Initialization:
 ARMA model:                arma
 Formula mean:              ~ arma(0, 0)
 GARCH model:               aparch
 Formula var:               ~ aparch(1, 1)
 ARMA Order:                0 0
 Max ARMA Order:            0
 GARCH Order:               1 1
 Max GARCH Order:           1
 Maximum Order:             1
 h.start:                   2
 llh.start:                 1
 Length of Series:          5037
 Recursion Init:            mci
 Series Scale:              1.000351

Parameter Initialization:
 Initial Parameters:          $params
 Limits of Transformations:   $U, $V
 Which Parameters are Fixed?  $includes
 Parameter Matrix:
                       U           V      params includes
    mu     -4.830854e-01   0.4830854 -0.04830854     TRUE
    omega   1.000703e-06 100.0703024  0.10007030     TRUE
    alpha1  1.000000e-08   1.0000000  0.10000000     TRUE
    gamma1 -1.000000e+00   1.0000000  0.10000000    FALSE
    beta1   1.000000e-08   1.0000000  0.80000000     TRUE
    delta   0.000000e+00   2.0000000  1.00000000    FALSE
    skew    1.000000e-01  10.0000000  1.00000000    FALSE
    shape   1.000000e+00  20.0000000  4.00000000    FALSE
 Index List of Parameters to be Optimized:
    mu  omega alpha1  beta1
     1      2      3      5
 Persistence:                  0.8797885

Iteration Path:



Now NLMINB


Error in gamma[i] : object is not subsettable
>






      ____________________________________________________________________________________

[[elided Yahoo spam]]
        [[alternative HTML version deleted]]

_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.

[John C. Frain]

Thanks.  The TS-Garch is now working as expected.

I had updated all the packages in the order specified except
fUnitroots.  When I load fGarch I get a warning message (output below)
which not be important (it is only a warning message).  The downloaded
fUnitroots.zip is only  9KB.  I have tried several times  but can not
download the full file.If I try

install.packages("fUnitRoots",repos="http://R-Forge.R-project.org")

I get a warning message

package  fUnitRoots is not available.

Best Regards

John

> library(fGarch)
Loading required package: fBasics
Loading required package: MASS
Loading required package: fImport
Loading required package: fSeries
Loading required package: robustbase
Loading required package: fCalendar
Loading required package: fEcofin
Loading required package: fUtilities
Rmetrics Package fUtilities (270.73) loaded.
Rmetrics Package fEcofin (270.73) loaded.
Rmetrics Package fCalendar (270.74) loaded.
Rmetrics Package fSeries (270.73) loaded.
Rmetrics Package fImport (270.73) loaded.
Rmetrics Package fBasics (270.73) loaded.
Loading required package: fArma
Rmetrics Package fArma (270.73) loaded.
Rmetrics Package fGarch (270.73) loaded.
Warning message:
In .recacheSubclasses(def@className, def, doSubclasses) :
  Undefined subclass, "double", of class "index_timeSeries";
definition not updated


2008/5/1 michal miklovic <mmiklovic@...>:

>
> Hi,
>
> it is a bug and it has already been solved in the development version of
> fGarch. Rmetrics development web is at:
> http://r-forge.r-project.org/projects/rmetrics
> where you can find the latest versions of all packages. Download the zipped
> Windows binaries and install them in the following order:
> fUtilities, fEcofin, fCalendar, fSeries, fImport, fBasics, fBonds, fArma,
> fGarch, fTrading, fExtremes, fNonlinear, fOptions, fAsianOptions,
> fExoticOptions, fUnitRoots, fMultivar, fCopulae, fRegression, fAssets,
> fPortfolio, Rmetrics
> and the estimation should work fine.
>
> Best regards,
>
> Michal
>
>
>
>
> ----- Original Message ----
> From: John Frain <frainj@...>
> To: R-SIG-Finance mailing list <r-sig-finance@...>
> Sent: Thursday, May 1, 2008 11:23:47 PM
> Subject: [R-SIG-Finance] Estimating the T-S Garch model
>
>  I am trying to estimate a T-S Garch model with the following code -
>
> library(fGarch)
> myFinCenter =  "GMT"
> dframe=read.csv(file="loss.csv")
> dframe[1:5,]
> loss=as.timeSeries(dframe)
> head(loss)
> tail(loss)
> fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0,
> include.delta=FALSE,trace=FALSE)
> summary(fit)
> fit2 = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0,
> include.delta=FALSE,leverage=FALSE)
>
> Output is at the end of the email.  I can estimate a model with
> nonzero leverage (gamma1 non-zero) but I have problems when I try to
> force gamma1 to be zero.  (See message at end of output.  Is there a
> problem or have I not understood something.
> I am using Windows XP,  R 2.7.0 and fGarch 260.72.  The data-set
> loss.csv is attached.
>
>
> --
> John C Frain
> Trinity College Dublin
> Dublin 2
> Ireland
> www.tcd.ie/Economics/staff/frainj/home.htm
> mailto:frainj@...
> mailto:frainj@...
>
>
> ############## OUTPUT ##############################
>
> > library(fGarch)
> > myFinCenter =  "GMT"
> > dframe=read.csv(file="loss.csv")
> > dframe[1:5,]
>           X      loss
> 1 1988-01-05 -3.2854337
> 2 1988-01-06 -2.6706190
> 3 1988-01-07  0.3662351
> 4 1988-01-08 -1.8030495
> 5 1988-01-11  1.1893279
> > loss=as.timeSeries(dframe)
> > head(loss)
>                 TS.1
> 1988-01-05 -3.2854337
> 1988-01-06 -2.6706190
> 1988-01-07  0.3662351
> 1988-01-08 -1.8030495
> 1988-01-11  1.1893279
> 1988-01-12 -0.1981119
> > tail(loss)
>                 TS.1
> 2008-01-24 -4.8668943
> 2008-01-25  0.7808780
> 2008-01-28  1.7767990
> 2008-01-29 -1.1739951
> 2008-01-30 -0.5321964
> 2008-01-31  1.3622478
> > fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0,
> include.delta=FALSE,trace=FALSE)
> > summary(fit)
>
> Title:
>  GARCH Modelling
>
> Call:
>  garchFit(formula = ~aparch(1, 1), data = loss@Data, delta = 1,
>     include.delta = FALSE, trace = FALSE)
>
> Mean and Variance Equation:
>  ~arma(0, 0) + ~aparch(1, 1)
>
> Conditional Distribution:
>  dnorm
>
> Coefficient(s):
>         mu      omega      alpha1      gamma1      beta1
> -0.0624601  0.0274496  0.0816745  -0.3396468  0.9100114
>
> Error Analysis:
>         Estimate  Std. Error  t value Pr(>|t|)
> mu    -0.062460    0.011934  -5.234 1.66e-07 ***
> omega  0.027450    0.005269    5.210 1.89e-07 ***
> alpha1  0.081675    0.009555    8.548  < 2e-16 ***
> gamma1 -0.339647    0.047851  -7.098 1.27e-12 ***
> beta1  0.910011    0.011797  77.139  < 2e-16 ***
> ---
> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
>
> Log Likelihood:
>  6630.546    normalized:  1.316368
>
> Standadized Residuals Tests:
>                                 Statistic p-Value
>  Jarque-Bera Test  R    Chi^2  5720.432  0
>  Shapiro-Wilk Test  R    W      NA        NA
>  Ljung-Box Test    R    Q(10)  102.3393  0
>  Ljung-Box Test    R    Q(15)  108.0036  4.440892e-16
>  Ljung-Box Test    R    Q(20)  118.3874  5.551115e-16
>  Ljung-Box Test    R^2  Q(10)  80.30754  4.369838e-13
>  Ljung-Box Test    R^2  Q(15)  86.34728  4.738099e-12
>  Ljung-Box Test    R^2  Q(20)  88.63123  1.28495e-10
>  LM Arch Test      R    TR^2  66.55288  1.405827e-09
>
> Information Criterion Statistics:
>       AIC      BIC      SIC      HQIC
> -2.630751 -2.624274 -2.630753 -2.628482
>
> Description:
>  Thu May 01 20:58:17 2008 by user: John C Frain
>
> > fit = garchFit(formula = ~ aparch(1,1), data=loss@Data, delta=1.0,
> include.delta=FALSE,leverage=FALSE)
>
> Series Initialization:
>  ARMA model:                arma
>  Formula mean:              ~ arma(0, 0)
>  GARCH model:              aparch
>  Formula var:              ~ aparch(1, 1)
>  ARMA Order:                0 0
>  Max ARMA Order:            0
>  GARCH Order:              1 1
>  Max GARCH Order:          1
>  Maximum Order:            1
>  h.start:                  2
>  llh.start:                1
>  Length of Series:          5037
>  Recursion Init:            mci
>  Series Scale:              1.000351
>
> Parameter Initialization:
>  Initial Parameters:          $params
>  Limits of Transformations:  $U, $V
>  Which Parameters are Fixed?  $includes
>  Parameter Matrix:
>                       U          V      params includes
>     mu    -4.830854e-01  0.4830854 -0.04830854    TRUE
>     omega  1.000703e-06 100.0703024  0.10007030    TRUE
>     alpha1  1.000000e-08  1.0000000  0.10000000    TRUE
>     gamma1 -1.000000e+00  1.0000000  0.10000000    FALSE
>     beta1  1.000000e-08  1.0000000  0.80000000    TRUE
>     delta  0.000000e+00  2.0000000  1.00000000    FALSE
>     skew    1.000000e-01  10.0000000  1.00000000    FALSE
>     shape  1.000000e+00  20.0000000  4.00000000    FALSE
>  Index List of Parameters to be Optimized:
>     mu  omega alpha1  beta1
>     1      2      3      5
>  Persistence:                  0.8797885
>
> Iteration Path:
>
>
>
> Now NLMINB
>
>
> Error in gamma[i] : object is not subsettable
> >
>
>
>  ________________________________
> Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it
> now.

--
John C Frain
Trinity College Dublin
Dublin 2
Ireland
www.tcd.ie/Economics/staff/frainj/home.htm
mailto:frainj@...
mailto:frainj@...

_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.

[Yohan Chalabi]

>>>> "MM" == michal miklovic <mmiklovic@...>
>>>> on Thu, 1 May 2008 14:58:29 -0700 (PDT)


   MM> it is a bug and it has already been solved in the
   MM> development version of fGarch. Rmetrics development web is
   MM> at: http://r-forge.r-project.org/projects/rmetrics
   MM> where you can find the latest versions of all
   MM> packages. Download the zipped Windows binaries and install
   MM> them in the following order:
   MM> fUtilities, fEcofin, fCalendar, fSeries, fImport, fBasics,
   MM> fBonds, fArma, fGarch, fTrading, fExtremes, fNonlinear,
   MM> fOptions, fAsianOptions, fExoticOptions, fUnitRoots,
   MM> fMultivar, fCopulae, fRegression, fAssets, fPortfolio,
   MM> Rmetrics
   MM> and the estimation should work fine.

Note you can use the script installRmetrics to install "fGarch"
development package with

source("http://rmetrics.R-Forge.R-project.org/installRmetrics.R")
installRmetrics("fGarch", repos="http://R-Forge.R-project.org")

Or any other Rmetrics development package...

regards,
Yohan

--
PhD student
Swiss Federal Institute of Technology
Zurich

www.ethz.ch
www.rmetrics.org

NOTE:
Rmetrics Workshop: http://www.rmetrics.org/meielisalp.htm
June 29th - July 3rd Meielisalp, Lake Thune, Switzerland

_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.

[John C. Frain]

This script does not appear to work for me.  I have already loaded
fGarch 270.73 and when I try to run the script followed by the call to
the 'iinstall' program i get the error message below. (I have ran the
script from the internet and also downloaded it to my PC and the
result is the same).  |Packages|Update packages| works in the windows
GUI.  My version of fUnitRoots is 260.72 and it will not update.  I
had previously tried to download the zipped file for fUnitRoots and
failed.  I tried again this evening and failed again.  (I have
downloaded all the remaining Rmetrics files at this stage).

Best regards

John


R version 2.7.0 (2008-04-22)
Copyright (C) 2008 The R Foundation for Statistical Computing
ISBN 3-900051-07-0

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

  Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> source("H:\\Archives\\software_econometric\\R\\R-2.7.0\\Rmetrics\\installRmetrics.R")
> installRmetrics("fGarch", repos="http://R-Forge.R-project.org")
Error in FUN("fGarch"[[1L]], ...) : subscript out of bounds
>


2008/5/6 Yohan Chalabi <chalabi@...>:

> >>>> "MM" == michal miklovic <mmiklovic@...>
>  >>>> on Thu, 1 May 2008 14:58:29 -0700 (PDT)
>
>
>    MM> it is a bug and it has already been solved in the
>    MM> development version of fGarch. Rmetrics development web is
>    MM> at: http://r-forge.r-project.org/projects/rmetrics
>    MM> where you can find the latest versions of all
>    MM> packages. Download the zipped Windows binaries and install
>    MM> them in the following order:
>    MM> fUtilities, fEcofin, fCalendar, fSeries, fImport, fBasics,
>    MM> fBonds, fArma, fGarch, fTrading, fExtremes, fNonlinear,
>    MM> fOptions, fAsianOptions, fExoticOptions, fUnitRoots,
>    MM> fMultivar, fCopulae, fRegression, fAssets, fPortfolio,
>    MM> Rmetrics
>    MM> and the estimation should work fine.
>
>  Note you can use the script installRmetrics to install "fGarch"
>  development package with
>
>  source("http://rmetrics.R-Forge.R-project.org/installRmetrics.R")
>  installRmetrics("fGarch", repos="http://R-Forge.R-project.org")
>
>  Or any other Rmetrics development package...
>
>  regards,
>  Yohan
>
>  --
>  PhD student
>  Swiss Federal Institute of Technology
>  Zurich
>
>  www.ethz.ch
>  www.rmetrics.org
>
>  NOTE:
>  Rmetrics Workshop: http://www.rmetrics.org/meielisalp.htm
>  June 29th - July 3rd Meielisalp, Lake Thune, Switzerland
>
>

--
John C Frain
Trinity College Dublin
Dublin 2
Ireland
www.tcd.ie/Economics/staff/frainj/home.htm
mailto:frainj@...
mailto:frainj@...

_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.

[Brian G. Peterson]

John Frain wrote:
> This script does not appear to work for me.  I have already loaded
> fGarch 270.73 and when I try to run the script followed by the call to
> the 'iinstall' program i get the error message below. (I have ran the
> script from the internet and also downloaded it to my PC and the
> result is the same).  |Packages|Update packages| works in the windows
> GUI.  My version of fUnitRoots is 260.72 and it will not update.  I
> had previously tried to download the zipped file for fUnitRoots and
> failed.  I tried again this evening and failed again.  (I have
> downloaded all the remaining Rmetrics files at this stage).

>> Yohan wrote:
<snip>
>>  Note you can use the script installRmetrics to install "fGarch"
>>  development package with
>>
>>  source("http://rmetrics.R-Forge.R-project.org/installRmetrics.R")
>>  installRmetrics("fGarch", repos="http://R-Forge.R-project.org")
>>
>>  Or any other Rmetrics development package...


The likely cause here is that the .ZIP files have failed to build on
R-forge for whatever reason.  Since John is using Windows, he won't
likely be compiling them from source.  Perhaps somebody can take a look
at the R-forge Windows build logs for these two packages and figure out
what the build issue on Windows is.

Regards,

    - Brian

_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.

[Jeff Ryan]

No solution from me, but I think this may have something to do with
the build issue:

http://www.nabble.com/R-Forge-SVN-repositories%3A-R-CMD-build-check-error-on-Windows-machines-to16942927.html

Jeff

On Tue, May 6, 2008 at 4:53 PM, Brian G. Peterson <brian@...> wrote:

> John Frain wrote:
>
> > This script does not appear to work for me.  I have already loaded
> > fGarch 270.73 and when I try to run the script followed by the call to
> > the 'iinstall' program i get the error message below. (I have ran the
> > script from the internet and also downloaded it to my PC and the
> > result is the same).  |Packages|Update packages| works in the windows
> > GUI.  My version of fUnitRoots is 260.72 and it will not update.  I
> > had previously tried to download the zipped file for fUnitRoots and
> > failed.  I tried again this evening and failed again.  (I have
> > downloaded all the remaining Rmetrics files at this stage).
> >
>
>
> >
> > > Yohan wrote:
> > >
> >
>  <snip>
>
>
> >
> > >  Note you can use the script installRmetrics to install "fGarch"
> > >  development package with
> > >
> > >  source("http://rmetrics.R-Forge.R-project.org/installRmetrics.R")
> > >  installRmetrics("fGarch", repos="http://R-Forge.R-project.org")
> > >
> > >  Or any other Rmetrics development package...
> > >
> >
>
>
>  The likely cause here is that the .ZIP files have failed to build on
> R-forge for whatever reason.  Since John is using Windows, he won't likely
> be compiling them from source.  Perhaps somebody can take a look at the
> R-forge Windows build logs for these two packages and figure out what the
> build issue on Windows is.
>
>  Regards,
>
>    - Brian
>
>
>
>  _______________________________________________
>  R-SIG-Finance@... mailing list
>  https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>  -- Subscriber-posting only.
>  -- If you want to post, subscribe first.
>

--
There's a way to do it better - find it.
Thomas A. Edison

_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.

[Yohan Chalabi]

>>>> "JF" == "John Frain" <frainj@...>
>>>> on Tue, 6 May 2008 22:28:30 +0100


   JF> his script does not appear to work for me.  I have already
   JF> loaded
   JF> fGarch 270.73 and when I try to run the script followed by
   JF> the call to
   JF> the 'iinstall' program i get the error message below. (I
   JF> have ran the
   JF> script from the internet and also downloaded it to my PC
   JF> and the
   JF> result is the same).  |Packages|Update packages| works in
   JF> the windows
   JF> GUI.  My version of fUnitRoots is 260.72 and it will not
   JF> update.  I
   JF> had previously tried to download the zipped file for
   JF> fUnitRoots and
   JF> failed.  I tried again this evening and failed again.  (I have
   JF> downloaded all the remaining Rmetrics files at this stage).

There is an issue on R-Forge server with Windows binary packages.
According to the log files, there is no gfortran compiler and all
packages with Fortran source can cannot be build.

But you can compile the packages from source as long as you have the
required tools installed.

in Rterm type:

source("http://rmetrics.R-Forge.R-project.org/installRmetrics.R")
installRmetrics("fGarch", repos="http://R-Forge.R-project.org", type =
"source", suggests = FALSE)

Note with 'suggests = FALSE' suggested packages will not be installed.

regards,
Yohan

--
PhD student
Swiss Federal Institute of Technology
Zurich

www.ethz.ch
www.rmetrics.org

NOTE:
Rmetrics Workshop: http://www.rmetrics.org/meielisalp.htm
June 29th - July 3rd Meielisalp, Lake Thune, Switzerland

_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.

[John C. Frain]

---------- Forwarded message ----------
From: John Frain <frainj@...>
Date: 2008/5/7
Subject: Re: [R-SIG-Finance] Estimating the T-S Garch model
To: Yohan Chalabi <chalabi@...>


Yohan

 This worked well.  I followed the instructions in Appendix E of the R
 Installation and Administration manual to download the Windows toolset
 and the MS HTML help workshop.  (I used the second url for the help
 workshop as the first may have moved to another address).  I have
 Latex already installed.  I did not install the Inno setup installer
 as I presume that this is not need for packages.  I created a batch
 file with one line

 set path=c:\Rtools\bin;c:\Rtools\perl\bin;c:\Rtools\mingw\bin;c:\progra~1\htmlhe~1;C:\Progra~1\R\R-2.7.0\bin;%path%

 to allow access to the tools. When I reboot my old path is then
 restored (I think).  I did not change anything in
 R_HOME/src/gnuwin32/MkRules.  The process appeared to work and was
 easier to implement than I had expected.

 Again Thanks to all

 John

 2008/5/7 Yohan Chalabi <chalabi@...>:


> >>>> "JF" == "John Frain" <frainj@...>
 >  >>>> on Tue, 6 May 2008 22:28:30 +0100
 >
 >
 >    JF> his script does not appear to work for me.  I have already
 >    JF> loaded
 >    JF> fGarch 270.73 and when I try to run the script followed by
 >    JF> the call to
 >    JF> the 'iinstall' program i get the error message below. (I
 >    JF> have ran the
 >    JF> script from the internet and also downloaded it to my PC
 >    JF> and the
 >    JF> result is the same).  |Packages|Update packages| works in
 >    JF> the windows
 >    JF> GUI.  My version of fUnitRoots is 260.72 and it will not
 >    JF> update.  I
 >    JF> had previously tried to download the zipped file for
 >    JF> fUnitRoots and
 >    JF> failed.  I tried again this evening and failed again.  (I have
 >    JF> downloaded all the remaining Rmetrics files at this stage).
 >
 >  There is an issue on R-Forge server with Windows binary packages.
 >  According to the log files, there is no gfortran compiler and all
 >  packages with Fortran source can cannot be build.
 >
 >  But you can compile the packages from source as long as you have the
 >  required tools installed.
 >
 >  in Rterm type:
 >
 >
 >  source("http://rmetrics.R-Forge.R-project.org/installRmetrics.R")
 >  installRmetrics("fGarch", repos="http://R-Forge.R-project.org", type =
 >  "source", suggests = FALSE)
 >
 >  Note with 'suggests = FALSE' suggested packages will not be installed.
 >
 >
 >
 >  regards,
 >  Yohan
 >
 >  --
 >  PhD student
 >  Swiss Federal Institute of Technology
 >  Zurich
 >
 >  www.ethz.ch
 >  www.rmetrics.org
 >
 >  NOTE:
 >  Rmetrics Workshop: http://www.rmetrics.org/meielisalp.htm
 >  June 29th - July 3rd Meielisalp, Lake Thune, Switzerland
 >
 >





--
 John C Frain
 Trinity College Dublin
 Dublin 2
 Ireland
 www.tcd.ie/Economics/staff/frainj/home.htm
 mailto:frainj@...
 mailto:frainj@...



--
John C Frain
Trinity College Dublin
Dublin 2
Ireland
www.tcd.ie/Economics/staff/frainj/home.htm
mailto:frainj@...
mailto:frainj@...

_______________________________________________
R-SIG-Finance@... mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only.
-- If you want to post, subscribe first.