Roger Newson writes (in part):
> I think I have discovered a rarely-encountered bug/feature in
> the Stata -graph- command. (I use the very latest Stata 7
> under Windows 98 SE.) It appears that, if there is only one
> observation in the selected sample, and -ylog- (or -xlog-) is
> specified, and the single y-value (or x-value) is in the open
> interval between one and zero, then -graph- complains of
> non-positive values.
The issue at hand here has to do with Stata setting an automatic
width for the axis when the user does not specify it. With only
one point, Stata chooses to set the axis width to be 1 unit about
the point, so when the single point is in the (0,1) interval, 0
becomes the lower graphing limit and, thus, the log transform
tries to operate on non-positive values.
However, even if I do specify the axis labels and/or scales, -graph- still
complains in the same way:
