R help - Sep 2003 - extracting monthly temperature data

I know that R is very advanced when it comes to DateTime handling. I am
not quite as advanced as R however.

I just downloaded a stupendously ugly dataset of hourly air temperature
from 1985 to 2003. It has a great many NAs. I want to extract mean,
median, max, and min monthly values.

So far I have read it in as object. Date and Time are factors and Temp
is an int.
> summary(temp.dat)
         Date             Time             Temp         
 01/01/1986:    24   03:00  :  6457   Min.   :  -22.00  
 01/01/1987:    24   04:00  :  6457   1st Qu.:   23.00  
 01/01/1988:    24   05:00  :  6457   Median :   34.00  
 01/01/1989:    24   06:00  :  6457   Mean   :   34.35  
 01/01/1990:    24   10:00  :  6457   3rd Qu.:   45.00  
 01/01/1992:    24   11:00  :  6457   Max.   :   89.00  
 (Other)   :154794   (Other):116196   NA's   :52178.00  

I then made a POSIXt object for the dates and times.
> date.time <- strptime(paste(temp.dat$Date, temp.dat$Time),
"%m/%d/%Y
%H:")
> class(date.time)
[1] "POSIXt"  "POSIXlt"
> date.time[1]
[1] "1985-10-01 01:00:00"


Now I'm stuck.

Questions: 

What's the best class for reuniting the temperature data with the dates?
A time series? A list?

Depending on the answer, what's the best way to extract, say, the mean
temperatures for Sepetmber?

Thanks in advance, Andy

What is "class(temp.dat)"?  Have you considered
"?data.frame"?  If
"temp.dat" is a data.frame, you can recombine date.time with Temp via

	  temp.dat$date.time <- date.time.

	  Then you can "rm(date.time)", because the above command makes a
copy
as a column of the data.frame temp.dat.

	  I have not been working with POSIXt, so I can't answer your question 
about how to get monthly means, other than to ask if you know about 
subscripting with a logical vector, "aggregate" and
"tapply"?

hope this helps.  spencer graves

Andy Bunn wrote:
> I know that R is very advanced when it comes to DateTime handling. I am
> not quite as advanced as R however.
> 
> I just downloaded a stupendously ugly dataset of hourly air temperature
> from 1985 to 2003. It has a great many NAs. I want to extract mean,
> median, max, and min monthly values.
> 
> So far I have read it in as object. Date and Time are factors and Temp
> is an int.
> 
> 
>>summary(temp.dat)
> 
>          Date             Time             Temp         
>  01/01/1986:    24   03:00  :  6457   Min.   :  -22.00  
>  01/01/1987:    24   04:00  :  6457   1st Qu.:   23.00  
>  01/01/1988:    24   05:00  :  6457   Median :   34.00  
>  01/01/1989:    24   06:00  :  6457   Mean   :   34.35  
>  01/01/1990:    24   10:00  :  6457   3rd Qu.:   45.00  
>  01/01/1992:    24   11:00  :  6457   Max.   :   89.00  
>  (Other)   :154794   (Other):116196   NA's   :52178.00  
> 
> I then made a POSIXt object for the dates and times.
> 
> 
>>date.time <- strptime(paste(temp.dat$Date, temp.dat$Time),
"%m/%d/%Y
> 
> %H:")
> 
>>class(date.time)
> 
> [1] "POSIXt"  "POSIXlt"
> 
>>date.time[1]
> 
> [1] "1985-10-01 01:00:00"
> 
> 
> Now I'm stuck.
> 
> Questions: 
> 
> What's the best class for reuniting the temperature data with the
dates?
> A time series? A list?
> 
> Depending on the answer, what's the best way to extract, say, the mean
> temperatures for Sepetmber?
> 
> Thanks in advance, Andy
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help