# read the data into the data.frame "x"

x = read.delim("clipboard")

# the "Year" column contains always 12 identical values in a row.

# -> create a time-series vector with 12 months per year:

Year = seq(

    as.Date(sprintf("%s/1/1",min(x$Year))),

    as.Date(sprintf("%s/12/1",max(x$Year))),

    "months")

# create a littel function to do the plotting for a given data set:

# Ithe function draws two polygons: a blue one for the positive values in y

# and a red one for the negative values in y

# values that should not be plottet are set to 0

myPlot = function(y, x,

    colors=c("blue","red"),

    xlab=deparse(substitute(x)),

    ylab=deparse(substitute(y)), ...)

{

    n = length(x)

    y.pos = y.neg = y

    y.pos[y0] = 0

    plot(y~x, type="n", xlab=xlab, ylab=ylab, ...)

    grid()

    polygon(c(x[1],x,x[n]),c(0,y.pos,0),col=colors[1],border=NA)

    polygon(c(x[1],x,x[n]),c(0,y.neg,0),col=colors[2],border=NA)

}

# now making the plots:

# two plots in one window, data from col. A in the upper plot, B in the lower:

par(mfcol=c(2,1),mar=c(4,4,1,1))

myPlot(x$A, Year, ylab="Column A values")

myPlot(x$B, Year, ylab="Column B values")

Look at the ggplot2 library in R.  Very powerful. 

# read the data into the data.frame "x"

x = read.delim("clipboard")

# the "Year" column contains always 12 identical values in a row.

# -> create a time-series vector with 12 months per year:

Year = seq(

    as.Date(sprintf("%s/1/1",min(x$Year))),

    as.Date(sprintf("%s/12/1",max(x$Year))),

    "months")

# create a littel function to do the plotting for a given data set:

# Ithe function draws two polygons: a blue one for the positive values in y

# and a red one for the negative values in y

# values that should not be plottet are set to 0

myPlot = function(y, x,

    colors=c("blue","red"),

    xlab=deparse(substitute(x)),

    ylab=deparse(substitute(y)), ...)

{

    n = length(x)

    y.pos = y.neg = y

    y.pos[y0] = 0

    plot(y~x, type="n", xlab=xlab, ylab=ylab, ...)

    grid()

    polygon(c(x[1],x,x[n]),c(0,y.pos,0),col=colors[1],border=NA)

    polygon(c(x[1],x,x[n]),c(0,y.neg,0),col=colors[2],border=NA)

}

# now making the plots:

# two plots in one window, data from col. A in the upper plot, B in the lower:

par(mfcol=c(2,1),mar=c(4,4,1,1))

myPlot(x$A, Year, ylab="Column A values")

myPlot(x$B, Year, ylab="Column B values")

and with ggplot2 library :

require(ggplot2)

ggplot(data=x, aes(x= Date))+    # Date as a date format

     geom_ribbon(aes(ymin= A, ymax= (A+B)/2,fill="blue") +

     geom_ribbon(aes(ymin= (A+B)/2, ymax= B),fill="red") +

     geom_line(aes(y=(A+B)/2), col = "black") +

     theme_bw()+

     facet_grid(group~.)                  # if there is a group

Hi Raoof,

I tried Denis' option but didn't worked and in fact I think it is not what you asked for.

I tried using geom_ribbon() but unfortunately you cant change the color of a part of the ribbon. One solution would be to plot two ribbons (which I tried but failed, Rplot01.png), the other is plotting segments with geom_segment() (see Rplot.png) which is not exactly what you want but at least is accurate.

good luck

qplot is a "quick plot" version of ggplot by the way.

 --------R script------------------

library(ggplot2)

library(readxl)

library(tidyr)

data

@ All, I do appreciate your time and valuable inputs.

@Mitchel, Thanks for the first tip on my query.

@Jochen, a straightforward direction

I've drawn the graph with your directions in a right way (attached).

@ Denis,

I tried your way, but didn't worked for me (may I missed something). Any way, appreciate your effort.

It made me encouraged to read that ggplot2 package.

@ Derek,

I did the Jochen`s command, successfully, Very straight commands (That was I want). I've spent a lot of time to do the Denis`s solution. I thought I had made a mistake somewhere.

Your proposed method was also interesting (especially the 2nd graph). Another way to visualization of the results!

I have to make some changes in your command to finalize the graph.

It was a valuable experience, AT LEAST FOR ME.

+ I have learned that the input data can be called from Excel (what I did not know).

Regards,

Raoof

Ok, you can create Derek's "segment plot" with base R in a very simple way, too:

par(mfcol=c(2,1),mar=c(4,4,1,1))

plot(x$A~Year, type="h", lwd=2, col=c("blue","red")[1+(x$A

Thanks for the following!

I did, but I got something like this (May something goes wrong-probably due to double introducing/assigning column names of the data set):

#cdm

library(ggplot2)

library(readxl)

library(tidyr)

data

Your problem is here:

    Year=data$Year

where Year is repeated 12 times ("Year" does not distingush between the months within a year). So when you plot

    plot(x$A~Year, type="h", lwd=2, col=c("blue","red")[1+(x$A

@Jochen, Thank you for debugging my command and useful tips !

data

Raoof,

in your last post you calculate "data$y2" and "Year", but for plotting you use only data$y2. There is no need to calculate both. You need to calculate just the one you use for plotting (since you used "data$y2" you can omit the calculation of "Year"):

    data

@Sofia,

Thank you.

Nice suggestion with graphical representation and extra examples, as suggested by our friends Denis and Derek in their commands.

Dear Ipsit,

Thanks for the interesting resources.