[R] AR(2) coefficient interpretation

Wed Dec 24 12:02:12 CET 2008

Stephen,

when I think about you're problem I'm a little worried as it should be is
very simple.
If you think of a the more straightforward AR(1) model y_t = a0 + b*y_t-1
the intercept is the the value y_t=a0 on the scatterplot of y_t vs. y_t-1.
For your first series labelled "a" for example the scatter plot (in Excel)
shows that this intercept computed using the linest function is about
57,000,000.

For the 2-d regression y_t on y_t-1 and y_t-2 there are only 9 data points
and the ls parameters are

  b1       b2         a0

(Embedded image moved to file: pic10548.jpg)

This gives quite  good fit to the data - the lm model in R should give the
same values (I havn't tried it).
Try it and see if the coeffs agree with arima - if not, it may be there's
something funny going on in arima in R.
So you want an AR model and you're unsure of arima use lm on lagged values.

Gerard

             "Stephen Oman"                                                
             <stephen.oman at gma                                             
             il.com>                                                    To 
                                       "Gerard M. Keogh"                   
             23/12/2008 13:40          <GMKeogh at justice.ie>                
                                                                        cc 

                                                                   Subject 
                                       Re: [R] AR(2) coefficient           
                                       interpretation                      

Hi Gerard,

Thank you for your reply. My point is even though the model is not
suitable, the intercept shouldn't be 5 point sth when the univariate data
is more than 100 million so I wonder whether my interpretation of those
coefficients are correct. Anyway, i have done the acf and pacf and it seems
AR(2) is the right model. Stephen

On Mon, Dec 22, 2008 at 8:33 AM, Gerard M. Keogh <GMKeogh at justice.ie>
wrote:
  12 obs isn't much for an ar model to work off!
  but in any event, did you check the acf of your data and did it
  geometrically decay after 2 steps to 0?
  If not the model is unsuitable.

  Gerard

              Stephen Oman
              <stephen.oman at gma
              il.com>                                                    To
              Sent by:                  r-help at r-project.org
              r-help-bounces at r-                                          cc
              project.org
                                                                    Subject
                                        [R]  AR(2) coefficient
              22/12/2008 15:06          interpretation

  I am a beginner in using R and I need help in the interpretation of AR
  result
  by R.  I used 12 observations for my AR(2) model and it turned out the
  intercept showed 5.23 while first and second AR coefficients showed 0.40
  and
  0.46. It is because my raw data are in million so it seems the intercept
  is
  too small and it doesn't make sense. Did i make any mistake in my code?
  My
  code is as follows:

  r<-read.table("data.txt", dec=",", header=T)
  attach(r)
  fit<-arima(a, c(2,0,0))

  Thank you for your help first.

  --
  View this message in context:
  http://www.nabble.com/AR%282%29-coefficient-interpretation-tp21129322p21129322.html

  Sent from the R help mailing list archive at Nabble.com.

  ______________________________________________
  R-help at r-project.org mailing list
  https://stat.ethz.ch/mailman/listinfo/r-help
  PLEASE do read the posting guide
  http://www.R-project.org/posting-guide.html
  and provide commented, minimal, self-contained, reproducible code.

  **********************************************************************************

  The information transmitted is intended only for the person or entity to
  which it is addressed and may contain confidential and/or privileged
  material. Any review, retransmission, dissemination or other use of, or
  taking of any action in reliance upon, this information by persons or
  entities other than the intended recipient is prohibited. If you received
  this in error, please contact the sender and delete the material from any
  computer.  It is the policy of the Department of Justice, Equality and
  Law Reform and the Agencies and Offices using its IT services to disallow
  the sending of offensive material.
  Should you consider that the material contained in this message is
  offensive you should contact the sender immediately and also
  mailminder[at]justice.ie.

  Is le haghaidh an duine nó an eintitis ar a bhfuil sí dírithe, agus le
  haghaidh an duine nó an eintitis sin amháin, a bheartaítear an fhaisnéis
  a tarchuireadh agus féadfaidh sé go bhfuil ábhar faoi rún agus/nó faoi
  phribhléid inti. Toirmisctear aon athbhreithniú, atarchur nó leathadh a
  dhéanamh ar an bhfaisnéis seo, aon úsáid eile a bhaint aisti nó aon
  ghníomh a dhéanamh ar a hiontaoibh, ag daoine nó ag eintitis seachas an
  faighteoir beartaithe. Má fuair tú é seo trí dhearmad, téigh i dteagmháil
  leis an seoltóir, le do thoil, agus scrios an t-ábhar as aon ríomhaire.
  Is é beartas na Roinne Dlí agus Cirt, Comhionannais agus Athchóirithe
  Dlí, agus na nOifígí agus na nGníomhaireachtaí a úsáideann seirbhísí TF
  na Roinne, seoladh ábhair cholúil a dhícheadú.
  Más rud é go measann tú gur ábhar colúil atá san ábhar atá sa
  teachtaireacht seo is ceart duit dul i dteagmháil leis an seoltóir
  láithreach agus le mailminder[ag]justice.ie chomh maith.
  ***********************************************************************************