[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: RE: seasonality
Your example and your explanation
show more clearly what is of interest to you.
I think this is now mainly a question for
people who work with this kind of data.
But I find it difficult to see how you
can test for seasonality without some time
series model of what else is going on, which
I don't see here.
I don't know what F-test you have in mind, but
classical anova is surely vitiated by serial dependence.
I can't add more.
> thank you for your reply. i express my seasonality question a
> bit clearer.
> suppose we are looking at five years of stock retruns data.
> how can we test
> for teh presence of seasonality in specific months?
> if i were to do the typical adjustment, i'd first calculated
> the average
> over teh entire time period, and divide my returns by that
> average. call the
> new variable rtn_adj. then i would have to average, by month,
> to get the
> seasonl adjsutement indices for each month (which should sum
> up to 12). i
> can then retruns and multiply teh original data by each of
> the appropriate
> seasonal adjsutemtn factor. My question: is there a simpler
> was to do this?
> also, how can i do f-test to test teh presence of
> seasonality? to give youa
> better idea, i'm attaching a sample of the data to give you
> an idea what i'm
> looking at.
> Date Returns
> December-04 0.072894168
> January-00 0.041277259
> >From: "Nick Cox" <email@example.com>
> >Reply-To: firstname.lastname@example.org
> >To: <email@example.com>
> >Subject: st: RE: seasonality
> >Date: Fri, 1 Apr 2005 16:57:48 +0100
> >Welcome back after your hiatus.
> >To centre a variable you need to go
> >su y, meanonly
> >gen yc = y - r(mean)
> >except that this is such a common need
> >that users have created commands to
> >do that for you. For example, look at Ben Jann's
> >-center- on SSC:
> >. ssc desc center
> >Your question about seasonality is much
> >fuzzier. There are lots of different ways
> >of testing for seasonality. In the
> >environmental sciences, I would usually
> >try fitting sine and cosine terms given
> >2 * pi * (position in year / length of year).
> >That is, also, I guess a congenial approach
> >for most natural scientists.
> >With economic or social data other methods appear
> >more common, and may or may not be more
> >appropriate. People seem happier with looking
> >at lags 4, 12, whatever depending on whether
> >data are quarterly, monthly, whatever. That may
> >be what SAS command proc x12 does. But
> >the adjustments seem much more complicated
> >given complications like holidays that are
> >irrelevant outside the human sphere.
> >In any case, graphics are often useful
> >for getting a handle on seasonality and
> >often surprising neglected by people like
> >Without knowing more about your data
> >or your research problem this is rather too
> >large a question to answer well. In short, the
> >question may be quick but the answer isn't.
> >kelly johnson
> > > I am returning to stata after a hatus of a couple of years. I
> > > had two quick questions:
> > >
> > > Suppose I am using a stream of time seies data for a single
> > > varible (Data, Variable1):
> > >
> > > (1) suppose i wanted to generate a new varible that equale
> > > variable1 - the
> > > mean of varible1. how do i do this (without having to create
> > > a whole column
> > > with only the mean of variable1 in it)?
> > >
> > > (2) is there an easy wasy to test for seasonality? in sas we
> > > have the proc
> > > x12 command? what's a quickand eqasy way to test for
> > > seasonality in data?
* For searches and help try: