R help - Jun 2012 - data analysis problem

Dear R users,

I have data on 4  types of interest rates. These rates evolve over
time and across regions of countries . So for each type  of interest
rates I want to run a regression of rates  on some other variables.
So my regression for one type of  interest rate will be I_{ij}_t= a
+regressors +error term.
where I_{ij}_t is the absolute difference in rates between two
locations i and j at time t. Note that i and j can be locations in
the same country or locations at different countries.
What I need  is construct a vector with all the pairs of locations for
a specific t. Put differently, I want to see how the interest rate
differential evolves over time for each pair of region. But the
monthly time series data I have available are heterogeneous across
countries

Take a look at the following table

  Country A                country B                  country C
country D     country E   country F

   '2-11-2002 '                07-12-2002'      
'23-11-2002'
'26-10-2002'    '27-12-2002'
.
.
.
09-10-2004'               '06-11-2004'              02-10-2004' 
09-10-2004'

>From the above table, In  country A the time starts at  "2/11/02 , in
country B the time starts at "07/12/02 and so forth.
Furthermore,  in  country A the time ends at  "9/10/04 , in country B
the time ends  at "06/11/02 and so forth.
As a result of this anomaly in the beginning of time, the time duration for
each country differs

So I cannot construct these pairs because for a particular time, t,
the rate exists in one location, but the rate in another location
starts after t or ends before t.

So the main thing I need to define is what I want done when data has
not yet started or is already finished in another country. I do not
know actually what the best solution is. This is my main question.
I found something about extrapolation (if this is to be the solution)
but I learn that extrapolation usually has quite a wide margin of
error!! Apart from that, I have no idea how to implement it in R.

Do you think that it would be better to try and create a more
symmetric sample so as the start and end dates across countries to be
very similar?
It is a data analysis problem. I need some help....

Stef:

1. Read and follow the posting guide. I could make no sense of your
post. This may be because I didn't work hard enough to decrypt it -
which I shouldn't have to do -- or because I'm too stupid -- which I
can't do anything about anyway.

2. What does this have to do with R anyway? Try posting on a
statistical list like stats.stackexchange.com if your primary concern
is "What should I do" rather than "How do I do _this_ in R?"

-- Bert

On Mon, Jun 4, 2012 at 5:11 PM, stef salvez <loggyedy at googlemail.com>
wrote:
> Dear R users,
>
> I have data on 4 ?types of interest rates. These rates evolve over
> time and across regions of countries . So for each type ?of interest
> rates I want to run a regression of rates ?on some other variables.
> So my regression for one type of ?interest rate will be I_{ij}_t= a
> +regressors +error term.
> where I_{ij}_t is the absolute difference in rates between two
> locations i and j at time t. Note that i and j can be locations in
> the same country or locations at different countries.
> What I need ?is construct a vector with all the pairs of locations for
> a specific t. Put differently, I want to see how the interest rate
> differential evolves over time for each pair of region. But the
> monthly time series data I have available are heterogeneous across
> countries
>
> Take a look at the following table
>
> ?Country A ? ? ? ? ? ? ? ?country B ? ? ? ? ? ? ? ? ?country C
> country D ? ? country E ? country F
>
> ? '2-11-2002 ' ? ? ? ? ? ? ? ?07-12-2002' ? ? ?
'23-11-2002'
> '26-10-2002' ? ?'27-12-2002'
> .
> .
> .
> 09-10-2004' ? ? ? ? ? ? ? '06-11-2004' ? ? ? ? ? ?
?02-10-2004' ?09-10-2004'
>
>
> >From the above table, In ?country A the time starts at ?"2/11/02 ,
in
> country B the time starts at "07/12/02 and so forth.
> Furthermore, ?in ?country A the time ends at ?"9/10/04 , in country B
> the time ends ?at "06/11/02 and so forth.
> As a result of this anomaly in the beginning of time, the time duration for
> each country differs
>
> So I cannot construct these pairs because for a particular time, t,
> the rate exists in one location, but the rate in another location
> starts after t or ends before t.
>
> So the main thing I need to define is what I want done when data has
> not yet started or is already finished in another country. I do not
> know actually what the best solution is. This is my main question.
> I found something about extrapolation (if this is to be the solution)
> but I learn that extrapolation usually has quite a wide margin of
> error!! Apart from that, I have no idea how to implement it in R.
>
> Do you think that it would be better to try and create a more
> symmetric sample so as the start and end dates across countries to be
> very similar?
> It is a data analysis problem. I need some help....
>
> ______________________________________________
> 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.


-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm