Skip to content

Problem with garch (tseries)

5 messages · Ricardo Zambrano Aguilera, Mark Leeds, Hannu Kahra +2 more

Original
Ricardo Zambrano Aguilera
# 
Greetings
i have this time series, (currency japanese log returns)
the data?s iid and it don?t have any serial correlation
but it does not fit to a regular garch ...
what could it be?
best regards

RZA
 <<JPY.txt>> 
pd:weekens removed

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: JPY.txt
Url: https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20060821/98166406/attachment.txt
Patrick Burns
# 
The Ljung-Box test is susceptible to outliers in
this context.  I haven't looked at this data at all,
but I suspect that a rank Ljung-Box test here
would give a dramatically different result.  More
details are in the Ljung-Box working paper on
the Burns Statistics website.


Patrick Burns
patrick at burns-stat.com
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")
Hannu Kahra wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    
2 days later

    

      
      

        


  

        

      Diethelm Wuertz
    

    

      
      #
    

  


  

      
Try:

    > require(fSeries)

    > x = scan("jpy.txt")
    > sd(x)
    [1] 0.005828643
   
    # The time series is badly scaled, use function scaling or just
    # multiply with 100!
   
    > x = 100 * x
    > sd(x)
    [1] 0.5828643


    > garchFit(series = x)
   
   
    Coefficient(s):
          mu         omega        alpha1         beta1 
    -8.88965e-05   1.06363e-02   1.94304e-02   9.48852e-01 
   
    Error Analysis:
             Estimate  Std. Error  t value Pr(>|t|)   
    mu     -0.0000889   0.0077757   -0.011   0.9909   
    omega   0.0106363   0.0052342    2.032   0.0421 * 
    alpha1  0.0194304   0.0079660    2.439   0.0147 * 
    beta1   0.9488518   0.0199498   47.562   <2e-16 ***
    ---
    Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
   
    Log Likelihood:
     1275.933    normalized:  0.8745259


Best regards Diethelm Wuertz

        
Ricardo Zambrano Aguilera wrote: