Combined plots with no margin

[Francois Durant]

Hi,

I would like to combine 3 plots with no margin (or as little as possible) and 1 unique legend that would not take (too much) vertical space. I'd like the graph to be divided in half on the vertical, with the first plot taking the first half of the vertical space. Then, I'd like the 2 other plots to be next to each other on the bottom part. Here is the data I have. I'm open to any package to use as long as it does the job

I'd like the 3 time periods to be the following:
- entire sample
- ym between [200 ; 250]
- ym between [225 ; 275]

For now, I run 3 time the code using keep to shrink the sample in the second and third plot as follow:
Code:
keep if ym>199 & ym<251
Here is the code I run 3 time with different intervals:

Code:
tsset ym keep if ym>199 & ym<251 line TRACAR ORCAR ym


Here is what I have so far, but it does not organize the graphs the way I described above:

 
 * Example generated by -dataex-. For more info, type help dataex clear input double(ym TRACAR ORCAR) 192 -.0033350477000837916 -.0003683217694234371 193   -.07045587821963419   -.06521660208867314 194   -.09191838589276918   -.08673469200194306 195   -.04931443098354621   -.03936599344240034 196   -.02715327012173765  -.020824142504374435 197  -.016842488342676973  -.008047950782653579 198  -.020084465205837943   -.01389383201526732 199  -.016233098943496338  -.021203868911942744 200 -.0035931515370219123   -.01828708093517132 201    .03571690334516507   .012819022862868242 202    .04455477981820734   .011654163595252115 203    .08485150376307166    .04743667762435774 204    .09984026051056252    .05645245801445384 205    .10889882174362192    .06250662555288229 206    .09408333428886506   .060906407025724876 207    .09724075886217153    .06934158203561133 208    .10286942199289811    .08089841744867378 209    .10209567085254202    .08651249161073829 210    .05754845029260604    .04469158442050881 211    .06638657070537657   .061282307727525034 212    .05460988250674513    .06586239601108772 213    .04152441848217898     .0688371759228358 214     .0438214294083841    .07872040889776405 215     .0674871893713507     .0977159595529987 216    .06490711716773598     .1174814795718476 217    .11062022619837128    .14999485112787375 218     .1359110146145715    .15769957980524812 219    .14543813812705447    .17252988410191014 220     .1483673419547573    .17973978385770514 221    .12099210505219944     .1740258272730565 222    .11255135129305215     .1735532944220967 223    .11492398701724553     .1910387372361875 224    .13497259566435454    .20138277593959808 225    .17795803505786037    .18837835235829675 226    .17746495255437128    .20494295565427204 227     .2094319405926391    .22772673635365004 228    .21822306291597904    .23514896897397897 229     .2548406367727308      .270475823730781 230    .28753585103992557     .3007595520490594 231    .28952136377471094    .30012048751904274 232      .305900475302077     .3142628384852584 233     .3219734184008402    .34054770817982427 234    .33806551259525347    .35550978509352427 235      .320176627847541     .3332741998516111 236     .3483171980612264     .3363191690086434 237     .3577985962232909    .35979190217196916 238     .3542809630115738    .35915773299828113 239     .3664674395147439     .3783511937582171 240     .3949898833795509    .41322580466915254 241    .39505513964152045     .3801585607353542 242     .3744761128443337    .38148197260561406 243     .3980570629159491     .4036312151654504 244     .4000157484726982     .4023994988784179 245     .4204809129467824    .41989629712318544 246     .4343275682730585     .4300761287219969 247     .4465203954056665    .43859389406875576 248     .4544472776790576     .4560526997400558 249     .4513158318481973      .457042252241649 250    .43428545338014796     .4529098012271615 251    .48364748671069896    .46851259940686074 252     .4891888644665549     .4853766770303783 253     .4806621036967232     .4802052187174724 254     .4909976207538399      .479991249739353 255     .4952901764776578    .48434629924774075 256      .502796695777313     .5110708866134684 257     .5287474544557784     .5338075577619873 258     .5260108421569143     .5316805781027812 259     .5501899064632975     .6093968954636813 260     .5176588528502183      .611569189925044 261    .48820828761561114      .574271605376266 262     .5053972122554777      .586187456312362 263     .5194708731016324     .5973020607573984 264    .49116969542143574     .5674053378361019 265     .5144010381475255     .5738188278093684 266     .5333427336933957     .5783666876254019 267     .5649628554518026     .6038820318286316 268     .5459778820914143     .5800562257899741 269      .573668638434915     .5903994846398564 270     .5652895376727657     .5883542205884337 271     .5831608234973406     .6015414719774771 272     .5902806810861686     .5984882409678033 273     .5774547977520452     .5985138514979731 274     .5902279186509635      .616901284979493 275     .5996878360196861      .625669798245332 276     .6134235367969734     .6366466304045999 277     .5963959663981883     .6160419151528446 278     .5965088885298631      .610828784838504 279     .6055140731272942     .6134342824420707 280     .6002134387534613     .6129732327585857 281     .6157646298670424      .612041277594986 282     .6239345752621089     .6154118313105509 283     .6183391080259679     .6179129335445327 284     .6256579297267545     .6197836663936542 285     .6265881542303138     .6102512514291559 286     .6238829846640681     .6103439922528602 287     .6366666986929777     .5979682539767863 288     .6345950622403422     .6190820881308037 289     .6404840963355422     .6103495384995195 290     .6601283352735469       .60406624306497 291      .667003569166827     .6142215186413874 end format %tm ym  preserve  line TRACAR ORCAR ym, saving(full, replace) xtitle("") restore  preserve keep if ym>199 & ym<251 line TRACAR ORCAR ym, lpattern(dash dash) saving(dotcom, replace) xtitle("") restore  preserve keep if ym>224 & ym<276 line TRACAR ORCAR ym, lpattern(dash dash) saving(GFC, replace) xtitle("") restore   grc1leg full.gph dotcom.gph GFC.gph, cols(1) /// imargin(0 0 0) legendfrom(GFC.gph)

Note that I have tried to post there, with no much success. I'm giving it another shot since I have really been struggling with this. Thanks a lot for your help in advance!
https://www.statalist.org/forums/for...d-in-1-subplot

[Tarun Choudhary]

Francois Durant The following code produces the graph below. Is this the result you wanted?

preserve  
line TRACAR ORCAR ym, saving(full, replace) xtitle("") legend(off) fysize(60) plotregion(margin(zero))
restore  
preserve
keep if ym>199 & ym<251
line TRACAR ORCAR ym, lpattern(dash dash) saving(dotcom, replace) xtitle("") fysize(17) legend(off) plotregion(margin(zero))
restore  
preserve
keep if ym>224 & ym<276
line TRACAR ORCAR ym, lpattern(dash dash) saving(GFC, replace) xtitle("") fysize(23) legend(pos(6) row(1)) plotregion(margin(zero))
restore
gr combine full.gph dotcom.gph GFC.gph,col(1) imargin(0 0 0) iscale(*0.7) xcommon

[Francois Durant]

Thanks a lot for your help Tarun! This is almost what I want. I'd like the 2 bottom graph to be next to each other on the bottom half of the graph (instead of being one below the other). Also, I'd like to keep only the dates where the graph spans for the 2 bottom graphs (around 1977 to 1980m5 for the first one). Finally, I'd like to get more frequent dates on the X-axis. That's a lot, but thank you so much for your help, again!

[Tarun Choudhary]

Francois Durant
The following code can do the first part asked by you. For the second part about the dates you should see axis_label_options and axis_label_options in the help file or the PDF documentation.

recode ym (200/250 = 1) (225/275 = 2) (else = .),gen(test)
line TRACAR ORCAR ym, saving(full, replace) xtitle("") legend(off) plotregion(margin(zero))
line TRACAR ORCAR ym, by(test, note("") ) lpattern(dash dash)saving(dotcom, replace) xtitle("") subtitle("") plotregion(margin(zero)) legend(pos(6) row(1)) plotregion(margin(zero)) 
gr combine full.gph dotcom.gph,col(1) imargin(0 0) iscale(*0.8)

Hope this helps.

[Francois Durant]

Hi Tarun,

It seems like the "recode" method does not work here because both periods are overlapping. Is there anything else you can suggest that would output the graphs on their respective interval only? Thanks, again.