help twoway qfit dialog: twoway
-------------------------------------------------------------------------------
Title
[G] graph twoway qfit -- Twoway quadratic prediction plots
Syntax
twoway qfit yvar xvar [if] [in] [weight] [, options]
options description
---------------------------------------------------------------------
range(# #) range over which predictions calculated
n(#) number of prediction points
atobs calculate predictions at xvar
estopts(regress_options) options for regress
predopts(predict_options) options for predict
cline_options change look of predicted line
axis_choice_options associate plot with alternative axis
twoway_options titles, legends, axes, added lines and
text, by, regions, name, aspect
ratio, etc.
---------------------------------------------------------------------
All options are rightmost; see repeated options.
aweights, fweights, and pweights are allowed. Weights, if specified,
affect estimation but not how the weighted results are plotted. See
weight.
Menu
Graphics > Twoway graph (scatter, line, etc.)
Description
twoway qfit calculates the prediction for yvar from a linear regression
of yvar on xvar and xvar^2 and plots the resulting curve.
Options
range(# #) specifies the x range over which predictions are calculated.
The default is range(. .), meaning the minimum and maximum values of
xvar. range(0 10) would make the range 0 to 10, range(. 10) would
make the range the minimum to 10, and range(0 .) would make the range
0 to the maximum.
n(#) specifies the number of points at which predictions over range() are
to be calculated. The default is n(100).
atobs is an alternative to n(). It specifies that the predictions be
calculated at the xvar values. atobs is the default if predopts() is
specified and any statistic other than xb is requested.
estopts(regress_options) specifies options to be passed along to regress
to estimate the linear regression from which the curve will be
predicted; see [R] regress. If this option is specified, commonly
specified is estopts(nocons).
predopts(predict_options) specifies options to be passed along to predict
to obtain the predictions after estimation by regress.
cline_options specify how the prediction line is rendered; see [G]
cline_options.
axis_choice_options associate the plot with a particular y or x axis on
the graph; see [G] axis_choice_options.
twoway_options are a set of common options supported by all twoway
graphs. These options allow you to title graphs, name graphs,
control axes and legends, add lines and text, set aspect ratios,
create graphs over by() groups, and change some advanced settings.
See [G] twoway_options.
Remarks
Remarks are presented under the following headings:
Typical use
Cautions
Use with by()
Typical use
twoway qfit is nearly always used in conjunction with other twoway
plottypes, such as
. sysuse auto
. scatter mpg weight || qfit mpg weight
(click to run)
Results are visually the same as typing
. generate tempvar = weight^2
. regress mpg weight tempvar
. predict fitted
. scatter mpg weight || line fitted weight
Cautions
Do not use twoway qfit when specifying the axis_scale_options yscale(log)
or xscale(log) to create log scales. Typing
. scatter mpg weight, xscale(log) || qfit mpg weight
produces something that is not a parabola because the regression
estimated for the prediction was for mpg on weight and weight^2, not mpg
on log(weight) and log(weight)^2.
Use with by()
qfit may be used with by() (as can all the twoway plot commands):
. scatter mpg weight || qfit mpg weight ||, by(foreign, total row(1))
(click to run)
Also see
Manual: [G] graph twoway qfit
Help: [G] graph twoway line, [G] graph twoway lfit, [G] graph twoway
fpfit, [G] graph twoway mband, [G] graph twoway mspline; [G]
graph twoway qfitci; [R] regress
