Skip to content

Unexpected Gap in simple line plot

5 messages · D Kelly O'Day, Bill Venables, Duncan Murdoch

D Kelly O'Day
# 
I am getting an unexpected gap in a simple plot of monthly data series.

I have created a csv file of monthly climate observations that I store
on-line. When I download the csv file and plot one of the series, I get a
gap even though there is data for the missing point.

Here is a snippet to show the problem.

  ## Strange plot results
    link <- "http://processtrends.com/files/RClimate_CTS_latest.csv"
    cts <- read.csv(link, header=T)

    ## Simple line plot - gap for no reason
    plot(cts$yr_frac, cts$GISS, type="l", xlim=c(1982, 1983),xaxs="i",
yaxs="i")

   ## May, 1982 observation missing
   ## Add points to plot in red, to see if May shows up
   points(cts$yr_frac, cts$GISS, type="p", col="red")
   ## yes, May shows up in points

## Look at cts data.frame. No obvious problems to me??
  small <- subset(cts, cts$yr_frac >1982 & cts$yr_frac <1983)
  small[,c(1,3)]

The same problem occurs in the other data vectors (HAD, NOAA, RSS, UAH, etc)

I have not been able to figure why I am getting the gap and how to show a
continuous line through May, 1982 data points.

Any suggestions?

D Kelly O'Day
http://chartsgraphs.worpress.com http://chartsgraphs.wordprss.com
Duncan Murdoch
#
On 20/01/2011 8:12 PM, D Kelly O'Day wrote:
Look at the data, not a subset of the data.  Printing cts, I see the 
following:

1226 1982.12 198202  0.07 -0.033  0.0801 -0.099 -0.201  0.1219  0.674 -0.052
1227 1982.21 198203 -0.12 -0.123  0.0265 -0.167 -0.347  0.1262  0.503 -0.052
1228    4.00     NA    NA     NA      NA     NA     NA      NA     NA     NA
1229 1982.29 198204 -0.03  0.024  0.1508 -0.116 -0.236  0.1311  0.495 -0.150
1230    5.00     NA    NA     NA      NA     NA     NA      NA     NA     NA
1231 1982.38 198205  0.10  0.041  0.1268 -0.228 -0.236  0.1730  0.876 -0.168

You get a better view in the data editor, seen by edit(cts) (on Windows, 
but I think it works on other systems too...)

There are a couple of weird records with NA values around the gap.  NA 
in a line plot causes a gap in the line.

Duncan Murdoch
Bill Venables
# 
You do have missing values.  Setting xlim does not subset the data. 

How about

########
link <- "http://processtrends.com/files/RClimate_CTS_latest.csv"
cts <- read.csv(link, header = TRUE)

scts <- subset(cts, !is.na(GISS) & !is.na(cts))   ## remove defectives

plot(GISS ~ yr_frac, scts, type = "l", xlim = c(1982, 1983),
     xaxs = "i", yaxs = "i")
points(GISS ~ yr_frac, scts, type = "p", col = "red")
########

?

Bill Venables
-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of D Kelly O'Day
Sent: Friday, 21 January 2011 11:12 AM
To: r-help at r-project.org
Subject: [R] Unexpected Gap in simple line plot


I am getting an unexpected gap in a simple plot of monthly data series.

I have created a csv file of monthly climate observations that I store
on-line. When I download the csv file and plot one of the series, I get a
gap even though there is data for the missing point.

Here is a snippet to show the problem.

  ## Strange plot results
    link <- "http://processtrends.com/files/RClimate_CTS_latest.csv"
    cts <- read.csv(link, header=T)

    ## Simple line plot - gap for no reason
    plot(cts$yr_frac, cts$GISS, type="l", xlim=c(1982, 1983),xaxs="i",
yaxs="i")

   ## May, 1982 observation missing
   ## Add points to plot in red, to see if May shows up
   points(cts$yr_frac, cts$GISS, type="p", col="red")
   ## yes, May shows up in points

## Look at cts data.frame. No obvious problems to me??
  small <- subset(cts, cts$yr_frac >1982 & cts$yr_frac <1983)
  small[,c(1,3)]

The same problem occurs in the other data vectors (HAD, NOAA, RSS, UAH, etc)

I have not been able to figure why I am getting the gap and how to show a
continuous line through May, 1982 data points.

Any suggestions?

D Kelly O'Day
http://chartsgraphs.worpress.com http://chartsgraphs.wordprss.com
D Kelly O'Day
# 
Bill & Duncan

Thanks for your quick reply. I would still be looking for days.

Now I have to figure out how the bad data got into cts since I generate this
file each month.
Duncan Murdoch
#
On 20/01/2011 9:33 PM, D Kelly O'Day wrote:
When I read that .csv file in OpenOffice, the lines with the NAs arise 
because the line before has an extra column.  That might be a hint as to 
what's going wrong in the generation...

Duncan Murdoch