How to plot two curves in the same Figure by Stata?

[Wei WANG]

Hello everyone,

I meet with an issue of plotting by Stata. I could you can help me.

There is a dependent variable A, an independent variable B, and a moderate variable C (C is also a dummy variable). Now, I need to get the scatterplot and its fitted curve of B*C and A, because C is a dummy variable, I want to get two curves in the same figure for C=1 and C=0 respectively.

How can I make this plot?

Thank you for answering, I really appreciate it.

[Marcos Almeida]

Welcome to the Stata Forum / Statalist.

Please read the FAQ, particularly on how to share data.

This is the best way to entail a helpful reply.

That said, hopefully the example below applies to your needs:

sysuse auto
scatter mpg price, by(foreign)

Edited to add the code and comment:

For interaction terms, you may try something like:

gen repprice = rep78*price
twoway( scatter mpg repprice if foreign ==0) (scatter mpg repprice if foreign == 1), by(foreign)

That said, it seems you need interaction terms plotted in the graph. If the model is, say, a regression analysis, you may use - margins - followed by marginspot.

[Nick Cox]

regplot can do this. There is no interaction term in this example, but there could be.

. sysuse auto, clear
(1978 Automobile Data)

. regress mpg weight foreign

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(2, 71)        =     69.75
       Model |   1619.2877         2  809.643849   Prob > F        =    0.0000
    Residual |  824.171761        71   11.608053   R-squared       =    0.6627
-------------+----------------------------------   Adj R-squared   =    0.6532
       Total |  2443.45946        73  33.4720474   Root MSE        =    3.4071

------------------------------------------------------------------------------
         mpg |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      weight |  -.0065879   .0006371   -10.34   0.000    -.0078583   -.0053175
     foreign |  -1.650029   1.075994    -1.53   0.130      -3.7955    .4954422
       _cons |    41.6797   2.165547    19.25   0.000     37.36172    45.99768
------------------------------------------------------------------------------

. regplot, sep(foreign) mcolor(orange blue) fitopts(lcolor(orange blue))

Where does regplot come from?

. search regplot, sj

Search of official help files, FAQs, Examples, SJs, and STBs

SJ-10-1 gr0009_1  . . . .  Software update for model diagnostic graph commands
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N. J. Cox
        (help anovaplot, indexplot, modeldiag, ofrtplot, ovfplot,
        qfrplot, racplot, rdplot, regplot, rhetplot, rvfplot2,
        rvlrplot, rvpplot2 if installed)
        Q1/10   SJ 10(1):164
        provides new command rbinplot for plotting means or medians
        of residuals by bins; provides new options for smoothing
        using restricted cubic splines; updates anova examples

SJ-4-4  gr0009  . . . . . . . . . . Speaking Stata: Graphing model diagnostics
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N. J. Cox
        (help anovaplot, indexplot, modeldiag, ofrtplot, ovfplot,
        qfrplot, racplot, rdplot, regplot, rhetplot, rvfplot2,
        rvlrplot, rvpplot2 if installed)
        Q4/04   SJ 4(4):449--475
        plotting diagnostic information calculated from residuals
        and fitted values from regression models with continuous
        responses

Note: there is another rdplot in existence. Mine was first, but at a wild guess the other one is used more, so in due course I will change the name of mine.

[Bruce Weaver]

Following up on the code Marcos Almeida posted in #2, you can use lfit and qfit to add linear and quadratic fit lines/curves. E.g.,

sysuse auto, clear
* Linear fit
scatter mpg price, by(foreign) || lfit mpg price, by(foreign)
* Quadratic fit
scatter mpg price, by(foreign) || qfit mpg price, by(foreign)
* Both linear and quadratic fits
scatter mpg price, by(foreign) || lfit mpg price, by(foreign)  || qfit mpg price, by(foreign)

HTH.

[Marcos Almeida]

Due to the paucity of information as well as lack of a workable data set, we have three members hazarding a guess.

As a tentative approach, I'm considering the use of - separate - so as to get the plot.

Also, perhaps something like this:

sysuse auto
gen highprice = price >= 6200
twoway (scatter mpg weight if highprice==0, legend(label(1 "Low Price"))) (scatter mpg weight if highprice ==1, legend(label(2 "High Price"))), by(foreign)

[Wei WANG]

Welcome to the Stata Forum / Statalist.

Please read the FAQ, particularly on how to share data.

This is the best way to entail a helpful reply.

That said, hopefully the example below applies to your needs:

sysuse auto
scatter mpg price, by(foreign)

[ATTACH=CONFIG]n1408617[/ATTACH]


Edited to add the code and comment:

For interaction terms, you may try something like:

gen repprice = rep78*price
twoway( scatter mpg repprice if foreign ==0) (scatter mpg repprice if foreign == 1), by(foreign)

That said, it seems you need interaction terms plotted in the graph. If the model is, say, a regression analysis, you may use - margins - followed by marginspot.

Thank you very much for your reply, your code is very useful to me. Thanks and best regards.

[Wei WANG]

Following up on the code Marcos Almeida posted in #2, you can use lfit and qfit to add linear and quadratic fit lines/curves. E.g.,

sysuse auto, clear
* Linear fit
scatter mpg price, by(foreign) || lfit mpg price, by(foreign)
* Quadratic fit
scatter mpg price, by(foreign) || qfit mpg price, by(foreign)
* Both linear and quadratic fits
scatter mpg price, by(foreign) || lfit mpg price, by(foreign) || qfit mpg price, by(foreign)

HTH.

Thank you sir, I know how to accomplish my task now. Thanks

[Selcuk Ozaydin]

I was wondering if it is possible to put different fits on different graphs. I want to display the linear and quadratic fits for the same data, is it possible to do that other than doing separately?

[Nick Cox]

Sure.

sysuse auto , clear
twoway scatter mpg weight || lfit mpg weight || qfit mpg weight
twoway scatter mpg weight || lfit mpg weight || qfit mpg weight , legend(order(2 "linear" 3 "quadratic") pos(1) ring(0) col(1)) ytitle("`: var label mpg'")

See also aaplot from SSC.

[Selcuk Ozaydin]

twoway scatter mpg weight || lfit mpg weight || qfit mpg weight This gives linear and quadratic fits on the same graph, I want them on different graphs. I couldn't make the second one work, I get an "invalid varlist" error Thanks for the quick response by the way.

[Nick Cox]

Looking at your original question in #8

(A) I was wondering if it is possible to put different fits on different graphs.

(B) I want to display the linear and quadratic fits for the same data, is it possible to do that other than doing separately?

I thought that you really meant (B), how to present graphs not separately, but in one. If you want them on separate graphs, what's the problem?

twoway scatter mpg weight || lfit mpg weight
twoway scatter mpg weight || qfit mpg weight

Are you asking about graph combine?

help graph combine

[Selcuk Ozaydin]

Yes! Thank you so much.