Skip to content

NAs for GARCH

3 messages · Rich Ghazarian, Brian G. Peterson, elton wang

#
Here is what my data looks like, after I try to remove the NAs, but I still get an error message for GARCH, any suggestion as to how I can index this.  The data is irregular daily returns, and I would like to fit a GARCH type model, but I am not sure how to apply zooreg or ts 

The raw data in CSV format


 
 
  2/5/2001
  0.29
 
 
  2/6/2001
  0.35
 
 
  2/7/2001
  0.31
 
 
  2/8/2001
  0.26
 
 
  2/9/2001
  0.26
 
 
  2/12/2001
  0.36
 
 
  2/13/2001
  0.43
 
 
  2/14/2001
  0.47
 
 
  2/15/2001
  0.34
 
 z1 <- read.zoo("pp.csv", sep = " ", format = "%m/%d/%y")

and apply
Time Series:

Start = 11358 

End = 13426 

Frequency = 1 

   [1]  0.291  0.354
 0.310  0.262  0.261  0.261  0.261  0.355
 0.427  0.473  0.340

  [12]  0.158  0.158
 0.158  0.158  0.021  0.032 -0.103 -0.103 -0.103 -0.103
-0.103

  [23] -0.208 -0.165  0.021
-0.078 -0.078 -0.078 -0.035 -0.020  0.065  0.065 -0.022

  [34] -0.022 -0.022 -0.024 -0.127
-0.127  0.000  0.152  0.152  0.152  0.464  0.392

  [45]  0.237  0.178
 0.069  0.069  0.069 -0.053  0.041  0.099  0.111
 0.157  0.157

  [56]  0.157  0.390
 0.397  0.172  0.179  0.068  0.068  0.068
 0.151  0.318  0.396

  [67]  0.452  0.452
 0.452  0.452  0.307  0.157  0.102 -0.120 -0.074
-0.074 -0.074

  [78]  0.069  0.028
 0.043  0.206  0.199  0.199  0.199  0.141
 0.066  0.085 -0.028

  [89] -0.053 -0.053 -0.053  0.068
 0.094  0.130  0.084 -0.050 -0.050 -0.050 -0.129

<!--
D(["mb"," [100] -0.147 -0.113  0.087\n 0.050  0.050  0.050  0.120  0.104  0.104\n-0.051 -0.148\u003c/font\>\n\u003cbr\>\n\u003cbr\>\u003cfont size\u003d\"2\" face\u003d\"sans-serif\"\>&gt; garch(z1, order \u003d c(1, 1), coef\n\u003d NULL, itmax \u003d 200, eps \u003d NULL,grad \u003d c(&quot;analytical&quot;,&quot;numerical&quot;),\nseries \u003d NULL, trace \u003d TRUE)\u003c/font\>\n\u003cbr\>\u003cfont size\u003d\"2\" face\u003d\"sans-serif\"\>Error in garch(z1, order \u003d c(1, 1),\ncoef \u003d NULL, itmax \u003d 200, eps \u003d NULL,  : \u003c/font\>\n\u003cbr\>\u003cfont size\u003d\"2\" face\u003d\"sans-serif\"\>        NAs in x\u003c/font\>\n\u003cbr\>\n--\u003d_alternative 007F353B8825729C_\u003d--",0]
);
D(["ma",[1,"\u003ctable class\u003datt cellspacing\u003d0 cellpadding\u003d5 border\u003d0\>\u003ctr\>\u003ctd colspan\u003d2\>\u003cb style\u003dpadding-left:3\>2 attachments\u003c/b\> &#8212; Scanning for viruses...\u003ctr\>\u003ctd\>\u003ctable cellspacing\u003d0 cellpadding\u003d0\>\u003ctr\>\u003ctd\>\u003cimg width\u003d16 height\u003d16 src\u003d\"/mail/images/generic.gif\"\>\u003ctd width\u003d7\>\u003ctd\>\u003cb\>gas.csv\u003c/b\>\u003cbr\>30K \u003c/table\>\u003ctr\>\u003ctd\>\u003ctable cellspacing\u003d0 cellpadding\u003d0\>\u003ctr\>\u003ctd\>\u003cimg width\u003d16 height\u003d16 src\u003d\"/mail/images/generic.gif\"\>\u003ctd width\u003d7\>\u003ctd\>\u003cb\>pp.csv\u003c/b\>\u003cbr\>23K \u003c/table\>\u003c/table\>","1114869f291fd79d"]
]
);

//--> [100] -0.147 -0.113  0.087
 0.050  0.050  0.050  0.120  0.104  0.104
-0.051 -0.148
= NULL, itmax = 200, eps = NULL,grad = c("analytical","numerical"),
series = NULL, trace = TRUE)

Error in garch(z1, order = c(1, 1),
coef = NULL, itmax = 200, eps = NULL,  : 

        NAs in x






--=_alternative 007F353B8825729C_=--

----- Original Message ----
From: "r-sig-finance-request at stat.math.ethz.ch" <r-sig-finance-request at stat.math.ethz.ch>
To: r-sig-finance at stat.math.ethz.ch
Sent: Monday, March 12, 2007 4:00:05 AM
Subject: R-SIG-Finance Digest, Vol 34, Issue 7

Send R-SIG-Finance mailing list submissions to
    r-sig-finance at stat.math.ethz.ch

To subscribe or unsubscribe via the World Wide Web, visit
    https://stat.ethz.ch/mailman/listinfo/r-sig-finance
or, via email, send a message with subject or body 'help' to
    r-sig-finance-request at stat.math.ethz.ch

You can reach the person managing the list at
    r-sig-finance-owner at stat.math.ethz.ch

When replying, please edit your Subject line so it is more specific
than "Re: Contents of R-SIG-Finance digest..."


Today's Topics:

   1. Re: Error in read.zoo (Achim Zeileis)


----------------------------------------------------------------------

Message: 1
Date: Sun, 11 Mar 2007 12:24:26 +0100 (CET)
From: Achim Zeileis <Achim.Zeileis at wu-wien.ac.at>
Subject: Re: [R-SIG-Finance] Error in read.zoo

Cc: r-sig-finance at stat.math.ethz.ch
Message-ID: <Pine.LNX.4.64.0703111221430.17202 at eowyn>
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
On Sat, 10 Mar 2007, Rich Ghazarian wrote:

            
Does this mean that there is no data? ;-)
More seriously, we would need an example to say anything useful.
Z
------------------------------

_______________________________________________
R-SIG-Finance mailing list
R-SIG-Finance at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-finance


End of R-SIG-Finance Digest, Vol 34, Issue 7
********************************************





 
____________________________________________________________________________________
Be a PS3 game guru.
#
On Monday 12 March 2007 20:42, Rich Ghazarian wrote:
As I said in an earlier post, you have a few options. zoo provides a 
number of functions which may help you decide what to do with missing 
days.  na.locf, na.omit,  na.approx, or thw aggregate function may be 
used.  You'll need to decide which method is most appropriate to your 
data and best represents your missing observations.  If enough 
observations are missing, for instance, you may need to use aggregate to 
fit a weekly regular series.  Only examination of your data will tell you 
what the right answer is, the list can be of limited assistance.

Regards,

   - Brian
#
This data set below is just missing weekends. On
businessday basis, they are actually continous. So
there is no point to go for irregular time frame
GARCH. 

irregular time frame will only be useful if there are
missing data, but from your data below, it looks to me
there are no missing data, there are no data at all on
those dates, this is a big difference.
--- Rich Ghazarian <rich7804 at yahoo.com> wrote:

            
-0.148\u003c/font\>\n\u003cbr\>\n\u003cbr\>\u003cfont
c(&quot;analytical&quot;,&quot;numerical&quot;),\nseries
width\u003d7\>\u003ctd\>\u003cb\>gas.csv\u003c/b\>\u003cbr\>30K
width\u003d7\>\u003ctd\>\u003cb\>pp.csv\u003c/b\>\u003cbr\>23K
----------------------------------------------------------------------
=== message truncated ===



 
____________________________________________________________________________________
8:00? 8:25? 8:40? Find a flick in no time