[R] Plotting numbers at a specified decimal length on a plot()
Marc Schwartz
marc_schwartz at comcast.net
Mon Sep 24 04:24:38 CEST 2007
On Sun, 2007-09-23 at 13:50 -0400, Matthew Dubins wrote:
> Hi there,
>
> I want to figure out how to plot means, with 2 decimal places, of any Y
> variable on a scatterplot according to any X variable (which obviously
> should have limited scope). I already figured out how to plot the
> means, but without limiting their precision to 2 decimal places. This
> is the code I used once I had the scatterplot drawn:
>
> text(c(1990, 1992, 1994, 1996, 1998, 1999, 2000, 2001, 2002, 2003), 25,
> mean.yrly.closures$values, cex=.7)
>
>
> It's a plot of school closures by Year. The closures variable is
> labeled "values" in the "mean.yrly.closures" data set. I managed to
> *display* the mean.yrly.closures data set with a precision of 2 decimal
> places (options(digits=2)), but then once I plotted the numbers, they
> showed up with many decimal places.
>
> It would be nice to figure this out so that I don't have to rely on
> Excel to make changes to the precision!
>
> Thanks,
> Matthew Dubins
R uses double precision floats internally to store numeric values by
default. See ?numeric
What you see displayed when a number is print()ed, is different by
default than the actual storage precision.
This is generally controlled by the use of options("digits"). However,
note that options("digits") does NOT control the number of digits to the
right of the decimal place, but the number of SIGNIFICANT digits
displayed. Also note that is it only a suggestion and not absolute. In
addition, when printing multiple values, such as a vector, other
characteristics will be in play. See ?print.default for more details
here.
To predictably control the number of digits to the right of the decimal
when displaying numbers, either in textual output or in a plot, you need
to format the output using functions such as sprintf() or formatC(), the
former being preferred. See ?sprintf and ?formatC for more information.
HTH,
Marc Schwartz
More information about the R-help
mailing list