[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: AW: RE: bar graph with 2 y-axes

From   "Martin Weiss" <>
To   <>
Subject   st: AW: RE: bar graph with 2 y-axes
Date   Mon, 27 Apr 2009 15:19:48 +0200


I was searching for an option to -jitter- (which some Stata -graph-s do
allow), but did not know you could introduce it through the back door :-)


-----Ursprüngliche Nachricht-----
[] Im Auftrag von Nick Cox
Gesendet: Montag, 27. April 2009 15:16
Betreff: st: RE: bar graph with 2 y-axes

One way to proceed is splendidly ad hoc. 

gen year = yofd(dofq(time))
collapse m1 consump , by(year)
gen year1 = year - 0.2
gen year2 = year + 0.2
twoway bar consump year1, barw(0.4) || bar  m1 year2, barw(0.4)

(In this example, no doubt there is a case for a different kind of
-collapse-, but it doesn't affect the graphical issue.) 

Such age-old trickery was written up in 

SJ-7-1  gr0026  . . . .  Stata tip 42: The overlay problem: Offset for
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J.
        Q1/07   SJ 7(1):141--142                                 (no
        tip for graphing several quantities on a continuous axis


Jochen Späth

After screening the statalist and some graphical advice literature in the
last days I still haven't got the clue: 
I need to draw a bar graph for two time series with different scales (e.g.
one in millions, the other in billions) over years. 
Graph bar delivers the desired over-all look as, f.i., in

-graph bar var1 var2,  over(year)-.

But I would like to have two different scales (left and right). 
-graph bar- does not offer such an option with the -over-option - logically.

In -twoway bar-, which has the option for a second vertical axis as in

-graph twoway (bar var1 year, yscale(1)) (bar var2 year, yscale(2))-, 

I have not managed to override the default result that the bars are drawn
one behind the other - I rather want them one beside the other (and grouped
over years). 

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index