R help - Feb 2005 - best analysis method : for time series ans cross sectional data

Howdy

What I 'd like to analyze with a large data on building permits is to find
time series effect of urban policy on buildings as well as
cross-sectional effects in any. In 1990 the specialZone urban policy
was introduced. I guess that the effects of this specialZone policy
would be different from countys. There are counties that do not
welcome this specialZone forced to design it.

One of the important aims is to find 1) time series effect using Dummy
variable,  2) cross-sectional effects using specialZones variable
below.

The data has items like year(1970-2000), floorSpace, county,
specialZones agianst permitting large buildings. specialZones have
been designed after 1990.
(Dummy = 1 after 1990, Dummy =0 before 1990)

I have tried three methods, such as
 lm(floorSpace ~ county, specialZones, Dummy), 
 glm(floorSpace ~ county, specialZones, Dummy),
 aov(floorSpace ~ county, specialZones, Dummy).

What I am focusing on is best method among lm, glm, aov or others not
siginificant results.

I have wasted  too  much time for this. I welcome your comments.

Thanks a lot,

-- 
Kum-Hoe Hwang, Ph.D.

Kyonggi Research Institute, Korea (ROK)
(Urban Planning and GIS)
Phone : 82-31-250-3283
Email : phdhwang at gmail.com

It looks to me like what you want is "intervention analysis" in 
the time series literature.  Have you considered the arima function, 
especially the example in the documentation using the xreg argument?  
Also, have you looked at ch. 14 in Venables and Ripley (2002) Modern 
Applied Statistics with S, 4th ed. (Springer)? 

      There are other time series packages available, e.g, dse, fSeries, 
its, GeneTS, msm, pastecs, splancs, tseries, urca, uroot, but I haven't 
used them and so can't comment further on them. 

      hope this helps.  spencer graves

Kum-Hoe Hwang wrote:
>Howdy
>
>What I 'd like to analyze with a large data on building permits is to
find
>time series effect of urban policy on buildings as well as
>cross-sectional effects in any. In 1990 the specialZone urban policy
>was introduced. I guess that the effects of this specialZone policy
>would be different from countys. There are counties that do not
>welcome this specialZone forced to design it.
>
>One of the important aims is to find 1) time series effect using Dummy
>variable,  2) cross-sectional effects using specialZones variable
>below.
>
>The data has items like year(1970-2000), floorSpace, county,
>specialZones agianst permitting large buildings. specialZones have
>been designed after 1990.
>(Dummy = 1 after 1990, Dummy =0 before 1990)
>
>I have tried three methods, such as
> lm(floorSpace ~ county, specialZones, Dummy), 
> glm(floorSpace ~ county, specialZones, Dummy),
> aov(floorSpace ~ county, specialZones, Dummy).
>
>What I am focusing on is best method among lm, glm, aov or others not
>siginificant results.
>
>I have wasted  too  much time for this. I welcome your comments.
>
>Thanks a lot,
>
>  
>

Hello, All:

	  What documentation do you recommend for someone trying to learn how 
to analyze time series in R beyond ch. 14 in Venables and Ripley (2002) 
Modern Applied Statistics with S, 4th ed. (Springer), and a not-quite 
random walk through the documentation on commands like arima and 
provided with packages like dse?  I've installed dse and its, and I've 
made progress using arima, acf, etc.  I tried working through 
"dse1-guide.pdf", but far I have not successfully used that package.

	  In particular, what's the preferred way to keep track of dates with 
time series?  I tried assigning a "Date" object somehow to a
"ts"
object, so far without success.  Two of my attempts are as follows:

 > tst2 <- ts(1:11, frequency=365,
+        start=c(2005, 11))
# This seemed to work, but the time does not seem to have class "Date"

 > tst3 <- ts(1:11, frequency=365,
+        start=as.Date("21/01/2005", "%d/%m/%Y"))
Error in Math.difftime((end - start) * frequency + 1.01) :
	floor not defined for difftime objects

	  Any suggestions would be greatly appreciated.  I'd gladly rtfm
("read
the f****** manual"), but I don't know which fm to r.

	  Thanks,
	  Spencer Graves	

############################
      It looks to me like what you want is "intervention analysis" in
the time series literature.  Have you considered the arima function,
especially the example in the documentation using the xreg argument?
Also, have you looked at ch. 14 in Venables and Ripley (2002) Modern
Applied Statistics with S, 4th ed. (Springer)?

      There are other time series packages available, e.g, dse, fSeries,
its, GeneTS, msm, pastecs, splancs, tseries, urca, uroot, but I haven't
used them and so can't comment further on them.

      hope this helps.  spencer graves

Kum-Hoe Hwang wrote:
>Howdy
>
>What I 'd like to analyze with a large data on building permits is to
find
>time series effect of urban policy on buildings as well as
>cross-sectional effects in any. In 1990 the specialZone urban policy
>was introduced. I guess that the effects of this specialZone policy
>would be different from countys. There are counties that do not
>welcome this specialZone forced to design it.
>
>One of the important aims is to find 1) time series effect using Dummy
>variable,  2) cross-sectional effects using specialZones variable
>below.
>
>The data has items like year(1970-2000), floorSpace, county,
>specialZones agianst permitting large buildings. specialZones have
>been designed after 1990.
>(Dummy = 1 after 1990, Dummy =0 before 1990)
>
>I have tried three methods, such as
> lm(floorSpace ~ county, specialZones, Dummy), 
> glm(floorSpace ~ county, specialZones, Dummy),
> aov(floorSpace ~ county, specialZones, Dummy).
>
>What I am focusing on is best method among lm, glm, aov or others not
>siginificant results.
>
>I have wasted  too  much time for this. I welcome your comments.
>
>Thanks a lot,
>
>  
>