Multiple plots and postscripts using split function

Hi

Maybe others will disagree but I find for cycle for this type of task better than sapply.

for(i in 1:length(ind)) {

if (there are more than 3 date items*) {

postscript(ind[i])
do all plotting
dev.off()
}}

If you want to plot with gaps you need to add all relevant YEARs for x axis with missing value in y before plotting. Then R plots it automatically with gap.

x<-1:10
y<-rnorm(10)
y[5:6]<-NA
plot(x,y, type="b")

I would suggest to use merge for this task.

table(dat$ID)
gives you number of unique values for each ID and you can use it for discarding this ID from the list.

Regards
Petr

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
project.org] On Behalf Of fd
Sent: Thursday, July 31, 2014 4:37 PM
To: r-help at r-project.org
Subject: [R] Multiple plots and postscripts using split function

Hi,

I'm relatively new to R and I would like to do the following:

I have a .csv file with four columns (NAME, ID, YEAR, VALUE) and would
like to do several xy plots with the year on the x-axis and the data
values
(measurements) on the y-axis and after that export the different plots
to postcript.

My .csv file looks something like this (only an example):

NAME                          ID              YEAR    VALUE
ADAMS                         885             1988            -2
ADAMS                         885             1989            0
BAHIA DEL DIABLO              2665            1999            4
BAHIA DEL DIABLO              2665            2000            8
BAHIA DEL DIABLO              2665            2001            19
BAHIA DEL DIABLO              2665            2002            13
BAHIA DEL DIABLO              2665            2003            13
BARTLEY                               893             1983            0
BARTLEY                               893             1984            -1
BARTLEY                               893             1985            0
BARTLEY                               893             1988            2
BARTLEY                               893             1989            -1
CANADA                                877             1972            -1

I have split the different items into groups and I'd like the plots to
have the title of NAME but the filename of the postscript to be
exported should have the ID as filename.

My code so far:

#Set Working Directory:
setwd("/Users/Desktop/FV")
# Read CSV
dat <- read.csv("FV.csv", sep=";", header=TRUE) # Split Data ind <-
split(x = dat,f = dat[,'ID']) nam <- names(ind)

sapply(nam, function(x) {
     postscript(x)
     par(mar=c(6,8,6,5), cex=0.8)
     plot(ind[[x]][,c('YEAR','VALUE')],
     type='b',
     main = x,
     xlab="Time [Years]",
     ylab="Front variation")
     axis(1, at = seq(1800,2100,5), cex.axis=1, labels=FALSE, tcl=-
0.3)
     axis(2, at = seq(-100000,100000,500), cex.axis=1, labels=FALSE,
tcl=-0.3)

     dev.off()
})

This results in plots with the title and filename of the resulting
postscript being the same. Is there a way to get the plot title out of
the NAME column and the filename out of the ID?

Additionally I'd only like to plot graphs for items with more than 3
data values. Is this possible to incorporate in the split command?

Another point is that some items have gaps in the time series where no
measurements were taken (in my example: BARTLEY from 1983 to 1985 and
1988 to 1989). I would like to plot using type= 'b' so that the points
are connected with lines, but when doing that, the values between 1985
and 1988 are automatically connected which I don't want. I'd like the
plot to start again at the value where the gap ends (in my example from
1988 onwards). Is there a solution for this?

Any help is kindly appreciated! Thanks for your help.

Kind regards,
fd



--
View this message in context: http://r.789695.n4.nabble.com/Multiple-
plots-and-postscripts-using-split-function-tp4694850.html
Sent from the R help mailing list archive at Nabble.com.

Thank you everyone for your help so far.

I am still working on the problem to get a merged new dataframe 

in new rows with NA values for each year that is missing for plotting 

gaps ( in the example the item BARTLEY: years 1984 to 1987 should 

with a row containing NA values). Could someone maybe help me 

Thank you everyone for your help so far.

I am still working on the problem to get a merged new dataframe which fills in new rows with NA values for each year that is missing for plotting with gaps ( in the example the item BARTLEY: years 1984 to 1987 should be filled with a row containing NA values). Could someone maybe help me with this? Thank you.


    NAME                                 ID             YEAR    VALUE
    ADAMS                               885             1988          -2
    ADAMS                               885             1989            0
    BAHIA DEL DIABLO    2665    1999            4
    BAHIA DEL DIABLO    2665    2000            8
    BAHIA DEL DIABLO    2665    2001          19
    BAHIA DEL DIABLO    2665    2002          13
    BAHIA DEL DIABLO    2665    2003          13
    BARTLEY                             893             1983            0
    BARTLEY                             893             1988            2
    BARTLEY                             893             1989           -1
    CANADA                              877             1972           -1
    CLARK CPI                   894             1973           -3

Am 01.08.2014 um 11:27 schrieb PIKAL Petr <petr.pikal at precheza.cz>:

Hi

Maybe others will disagree but I find for cycle for this type of task better than sapply.

for(i in 1:length(ind)) {

if (there are more than 3 date items*) {

postscript(ind[i])
do all plotting
dev.off()
}}

If you want to plot with gaps you need to add all relevant YEARs for x axis with missing value in y before plotting. Then R plots it automatically with gap.

x<-1:10
y<-rnorm(10)
y[5:6]<-NA
plot(x,y, type="b")

I would suggest to use merge for this task.

table(dat$ID)
gives you number of unique values for each ID and you can use it for discarding this ID from the list.

Regards
Petr

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
project.org] On Behalf Of fd
Sent: Thursday, July 31, 2014 4:37 PM
To: r-help at r-project.org
Subject: [R] Multiple plots and postscripts using split function

Hi,

I'm relatively new to R and I would like to do the following:

I have a .csv file with four columns (NAME, ID, YEAR, VALUE) and would
like to do several xy plots with the year on the x-axis and the data
values
(measurements) on the y-axis and after that export the different plots
to postcript.

My .csv file looks something like this (only an example):

NAME                          ID              YEAR    VALUE
ADAMS                         885             1988            -2
ADAMS                         885             1989            0
BAHIA DEL DIABLO              2665            1999            4
BAHIA DEL DIABLO              2665            2000            8
BAHIA DEL DIABLO              2665            2001            19
BAHIA DEL DIABLO              2665            2002            13
BAHIA DEL DIABLO              2665            2003            13
BARTLEY                               893             1983            0
BARTLEY                               893             1984            -1
BARTLEY                               893             1985            0
BARTLEY                               893             1988            2
BARTLEY                               893             1989            -1
CANADA                                877             1972            -1

I have split the different items into groups and I'd like the plots to
have the title of NAME but the filename of the postscript to be
exported should have the ID as filename.

My code so far:

#Set Working Directory:
setwd("/Users/Desktop/FV")
# Read CSV
dat <- read.csv("FV.csv", sep=";", header=TRUE) # Split Data ind <-
split(x = dat,f = dat[,'ID']) nam <- names(ind)

sapply(nam, function(x) {
     postscript(x)
     par(mar=c(6,8,6,5), cex=0.8)
     plot(ind[[x]][,c('YEAR','VALUE')],
     type='b',
     main = x,
     xlab="Time [Years]",
     ylab="Front variation")
     axis(1, at = seq(1800,2100,5), cex.axis=1, labels=FALSE, tcl=-
0.3)
     axis(2, at = seq(-100000,100000,500), cex.axis=1, labels=FALSE,
tcl=-0.3)

     dev.off()
})

This results in plots with the title and filename of the resulting
postscript being the same. Is there a way to get the plot title out of
the NAME column and the filename out of the ID?

Additionally I'd only like to plot graphs for items with more than 3
data values. Is this possible to incorporate in the split command?

Another point is that some items have gaps in the time series where no
measurements were taken (in my example: BARTLEY from 1983 to 1985 and
1988 to 1989). I would like to plot using type= 'b' so that the points
are connected with lines, but when doing that, the values between 1985
and 1988 are automatically connected which I don't want. I'd like the
plot to start again at the value where the gap ends (in my example from
1988 onwards). Is there a solution for this?

Any help is kindly appreciated! Thanks for your help.

Kind regards,
fd



--
View this message in context: http://r.789695.n4.nabble.com/Multiple-
plots-and-postscripts-using-split-function-tp4694850.html
Sent from the R help mailing list archive at Nabble.com.