Hi,
May be this is what you meant.
dat1<- read.csv("data2.csv",header=FALSE,sep="\t")
library(plyr)
res1<- ddply(dat1,.(V2,V3,V4),summarize,V5=round(mean(V5),2))
?head(res1)
#??? V2 V3 V4?? V5
#1 1961? 1? 1 0.52
#2 1961? 1? 2 0.00
#3 1961? 1? 3 0.00
#4 1961? 1? 4 0.00
#5 1961? 1? 5 0.12
#6 1961? 1? 6 0.12
?tail(res1)
#??????? V2 V3 V4?? V5
#16431 2005 12 26 0.37
#16432 2005 12 27 0.06
#16433 2005 12 28 0.05
#16434 2005 12 29 0.34
#16435 2005 12 30 0.84
#16436 2005 12 31 1.14
#or
res<- aggregate(V5~V2+V3+V4,data=dat1,FUN=mean)
res2<- res[order(res$V2,res$V3,res$V4),]
row.names(res2)<- 1:nrow(res2)
res2$V5<- round(res2$V5,2)
?identical(res1,res2)
#[1] TRUE
A.K.
----- Original Message -----
From: "Zilefac, Elvis" <aez849 at mail.usask.ca>
To: arun <smartpink111 at yahoo.com>
Cc:
Sent: Thursday, May 16, 2013 12:05 PM
Subject: RE: Loop through a simulation
Hi AK,
The output looks? a bit different from what I am expecting. The output format
should be the same as simulation 1
head(data2)
? V1? V2 V3 V4? V5
1? 1 1961? 1? 1 0.51
2? 1 1961? 1? 2 0.00
3? 1 1961? 1? 3 0.00
4? 1 1961? 1? 4 0.00
5? 1 1961? 1? 5 0.00
6? 1 1961? 1? 6 0.00
tail(data)
? V1? V2 V3 V4? V5
21 2005 12 26 0.00
21 2005 12 27 0.59
21 2005 12 28 0.00
21 2005 12 29 0.56
21 2005 12 30 3.70
21 2005 12 31 0.00
The output file should resemble the above.
You can arrive at the above output by sorting the first rows of sim 1, sim
2,...sim 21 and then take their mean.
Similarly, take the second row in sim 1, sim 2, ..., sim 21 and take their mean.
This should give you the output above.
Thank so much.
____________
From: arun [smartpink111 at yahoo.com]
Sent: Thursday, May 16, 2013 1:16 AM
To: Zilefac, Elvis
Cc: R help
Subject: Re: Loop through a simulation
HI,
Try this:
dat1<- read.csv("data2.csv",header=FALSE,sep="\t")
? dim(dat1)
#[1] 345156? ? ? 5
res<- aggregate(V5~V2+V3+V4,data=dat1,FUN=mean)
dim(res)
#[1] 16436? ? 4
head(res)
#? ? V2 V3 V4? ? ? ? V5
#1 1961? 1? 1 0.51761905
#2 1962? 1? 1 0.04666667
#3 1963? 1? 1 1.14761905
#4 1964? 1? 1 0.05190476
#5 1965? 1? 1 0.51428571
#6 1966? 1? 1 0.50952381
tail(res)
#? ? ? ? V2 V3 V4? ? ? ? V5
#16431 2000 12 31 0.6428571
#16432 2001 12 31 0.1600000
#16433 2002 12 31 0.3314286
#16434 2003 12 31 0.2271429
#16435 2004 12 31 3.7423810
#16436 2005 12 31 1.1371429
A.K.
----- Original Message -----
From: "Zilefac, Elvis" <aez849 at mail.usask.ca>
To: arun <smartpink111 at yahoo.com>
Cc:
Sent: Thursday, May 16, 2013 2:25 AM
Subject: RE: Loop through a simulation
Dear AK,
Here is data from 21 simulations or so.
Thanks so much.
________________________________________
From: arun [smartpink111 at yahoo.com]
Sent: Thursday, May 16, 2013 12:04 AM
To: Zilefac, Elvis
Subject: Re: Loop through a simulation
Dear Zilefac,
I am not able to download your dataset.? It says "An error occured while
downloading".? Could you send a smaller? dataset?? For example the data
between 1961-1965 or so.
________________________________
From: "Zilefac, Elvis" <aez849 at mail.usask.ca>
To: "smartpink111 at yahoo.com" <smartpink111 at yahoo.com>
Sent: Thursday, May 16, 2013 1:05 AM
Subject: Loop through a simulation
Dear AK,
Please help again. I have this data file to loop through. The columns are as
follows:
[simulation, year, month, day, rainfall_values]
I have daily data from 1961 to 2005 . I have made 63 stochastic simulations each
from 1961 to 2005.
Now, I would like to obtain the values for each rainfall_value, which is the
mean from 63 such values.
For example, rainfall_value of row one should be the mean(simulation 1 row 1,
simulation 2 row 1,...simulation? 63 row 1) etc. In the end, you will have daily
data from 1961 to 200, same length as simulation 1 ,for example.
Thanks so much for this great help.
Atem.