A Problem with connected plot

[Nihat Mugurtay]

Dear all,

I have a dataset below, composed of the region, year, and value. Normally, when I make a connected graph, I have each region's value as an independent column, and the code is like this:

twoway (connected asia year)(connected Africa year)(connected europe year)

However, in my dataset below, I don't want to copy-paste and create new columns manually, instead, I want to automatically generate a connected graph, based on the data available below. Or is there a handy way of automatically generating new columns based on each region's value for specific years? (without copy-pasting)

Thank you so much,

Best,

* Example generated by -dataex-. For more info, type help dataex
clear
input str22 regions int year float percent
"Africa"                 2002  85.17632
"Africa"                 2003  79.51936
"Africa"                 2004  83.43472
"Africa"                 2005  46.13064
"Africa"                 2006  44.20268
"Africa"                 2007  46.07319
"Africa"                 2008  44.70668
"Africa"                 2009  43.12626
"Africa"                 2010  43.63366
"Africa"                 2011  36.20064
"Africa"                 2012  41.24461
"Africa"                 2013 37.146824
"Africa"                 2014  39.44272
"Africa"                 2015 36.002865
"Africa"                 2016 36.622124
"Africa"                 2017  36.33076
"Africa"                 2018 37.235317
"Africa"                 2019  42.04245
"Africa"                 2020  40.90646
"America"                2002  3.500117
"America"                2003  3.934545
"America"                2004  5.955437
"America"                2005  8.889876
"America"                2006  8.688569
"America"                2007   9.34407
"America"                2008  8.561049
"America"                2009  8.548162
"America"                2010 10.196347
"America"                2011  7.605177
"America"                2012  6.895785
"America"                2013  6.745145
"America"                2014  5.685211
"America"                2015  7.563207
"America"                2016  6.287817
"America"                2017  7.283136
"America"                2018  7.202162
"America"                2019  5.270456
"America"                2020   6.00174
"Asia (w/o Middle-East)" 2002         0
"Asia (w/o Middle-East)" 2003 1.4860897
"Asia (w/o Middle-East)" 2004         0
"Asia (w/o Middle-East)" 2005 12.642322
"Asia (w/o Middle-East)" 2006 12.319653
"Asia (w/o Middle-East)" 2007 11.189074
"Asia (w/o Middle-East)" 2008 12.417957
"Asia (w/o Middle-East)" 2009 12.358888
"Asia (w/o Middle-East)" 2010 12.593313
"Asia (w/o Middle-East)" 2011  11.73253
"Asia (w/o Middle-East)" 2012  9.272342
"Asia (w/o Middle-East)" 2013  9.475549
"Asia (w/o Middle-East)" 2014  8.578097
"Asia (w/o Middle-East)" 2015  11.97774
"Asia (w/o Middle-East)" 2016 11.692198
"Asia (w/o Middle-East)" 2017 12.285928
"Asia (w/o Middle-East)" 2018 12.449587
"Asia (w/o Middle-East)" 2019 10.721766
"Asia (w/o Middle-East)" 2020  11.89148
"Europe"                 2002         .
"Europe"                 2003 1.2007693
"Europe"                 2004         0
"Europe"                 2005  12.30139
"Europe"                 2006  13.46489
"Europe"                 2007 12.825014
"Europe"                 2008  13.91023
"Europe"                 2009 17.804235
"Europe"                 2010 15.739152
"Europe"                 2011  31.67814
"Europe"                 2012  30.99681
"Europe"                 2013  31.76924
"Europe"                 2014  31.45511
"Europe"                 2015 28.392185
"Europe"                 2016  29.45614
"Europe"                 2017 26.051006
"Europe"                 2018  22.60656
"Europe"                 2019  21.57816
"Europe"                 2020 23.611736
"Middle East"            2002         0
"Middle East"            2003 .59221643
"Middle East"            2004         0
"Middle East"            2005  5.858264
"Middle East"            2006  9.175097
"Middle East"            2007  7.451975
"Middle East"            2008  7.945778
"Middle East"            2009  6.374172
"Middle East"            2010  6.231583
"Middle East"            2011 4.4821134
"Middle East"            2012  4.335175
"Middle East"            2013  7.054042
"Middle East"            2014  6.882815
"Middle East"            2015  7.979837
"Middle East"            2016   8.19168
"Middle East"            2017  7.772645
"Middle East"            2018  8.790713
"Middle East"            2019  8.470904
"Middle East"            2020  7.209192
"Ocenia"                 2002         0
"Ocenia"                 2003 2.0340345
"Ocenia"                 2004 2.4081006
"Ocenia"                 2005 1.0007294
"Ocenia"                 2006  .8274303
end

------------------ copy up to and including the previous line ------------------

Listed 100 out of 114 observations
Use the count() option to list more

[Ken Chui]

It's possible to use "if" to achieve that without creating new variables. Here is an example:

twoway connected percent year if regions == "Africa" || ///
       connected percent year if regions == "America" || ///
       connected percent year if regions == "Asia (w/o Middle-East)" || ///
       connected percent year if regions == "Europe" || ///
       connected percent year if regions == "Middle East" || ///
       connected percent year if regions == "Ocenia", ///
       legend(label (1 "Afr") label (2 "Am")  label (3 "Asia") ///
       label (4 "Eur") label (5 "Mid East") label (6 "Oce") col(3))

And just for completion, I'd like to address the original question as well. Yes, it's possible to spread the percent across. There are multiple ways. Here is a "wrong" use of reshape wide:

encode regions, gen(nreg)
reshape wide percent, i(regions year) j(nreg)

[Nick Cox]

Ken Chui gave a nice solution. Here are some more ideas. separate can be useful, but in this case the ensuing graph is fairly horrible.. so not shown here.

I've corrected a typo in the dataset.

* Example generated by -dataex-. For more info, type help dataex
clear
input str22 regions int year float percent
"Africa"                 2002  85.17632
"Africa"                 2003  79.51936
"Africa"                 2004  83.43472
"Africa"                 2005  46.13064
"Africa"                 2006  44.20268
"Africa"                 2007  46.07319
"Africa"                 2008  44.70668
"Africa"                 2009  43.12626
"Africa"                 2010  43.63366
"Africa"                 2011  36.20064
"Africa"                 2012  41.24461
"Africa"                 2013 37.146824
"Africa"                 2014  39.44272
"Africa"                 2015 36.002865
"Africa"                 2016 36.622124
"Africa"                 2017  36.33076
"Africa"                 2018 37.235317
"Africa"                 2019  42.04245
"Africa"                 2020  40.90646
"America"                2002  3.500117
"America"                2003  3.934545
"America"                2004  5.955437
"America"                2005  8.889876
"America"                2006  8.688569
"America"                2007   9.34407
"America"                2008  8.561049
"America"                2009  8.548162
"America"                2010 10.196347
"America"                2011  7.605177
"America"                2012  6.895785
"America"                2013  6.745145
"America"                2014  5.685211
"America"                2015  7.563207
"America"                2016  6.287817
"America"                2017  7.283136
"America"                2018  7.202162
"America"                2019  5.270456
"America"                2020   6.00174
"Asia (w/o Middle-East)" 2002         0
"Asia (w/o Middle-East)" 2003 1.4860897
"Asia (w/o Middle-East)" 2004         0
"Asia (w/o Middle-East)" 2005 12.642322
"Asia (w/o Middle-East)" 2006 12.319653
"Asia (w/o Middle-East)" 2007 11.189074
"Asia (w/o Middle-East)" 2008 12.417957
"Asia (w/o Middle-East)" 2009 12.358888
"Asia (w/o Middle-East)" 2010 12.593313
"Asia (w/o Middle-East)" 2011  11.73253
"Asia (w/o Middle-East)" 2012  9.272342
"Asia (w/o Middle-East)" 2013  9.475549
"Asia (w/o Middle-East)" 2014  8.578097
"Asia (w/o Middle-East)" 2015  11.97774
"Asia (w/o Middle-East)" 2016 11.692198
"Asia (w/o Middle-East)" 2017 12.285928
"Asia (w/o Middle-East)" 2018 12.449587
"Asia (w/o Middle-East)" 2019 10.721766
"Asia (w/o Middle-East)" 2020  11.89148
"Europe"                 2002         .
"Europe"                 2003 1.2007693
"Europe"                 2004         0
"Europe"                 2005  12.30139
"Europe"                 2006  13.46489
"Europe"                 2007 12.825014
"Europe"                 2008  13.91023
"Europe"                 2009 17.804235
"Europe"                 2010 15.739152
"Europe"                 2011  31.67814
"Europe"                 2012  30.99681
"Europe"                 2013  31.76924
"Europe"                 2014  31.45511
"Europe"                 2015 28.392185
"Europe"                 2016  29.45614
"Europe"                 2017 26.051006
"Europe"                 2018  22.60656
"Europe"                 2019  21.57816
"Europe"                 2020 23.611736
"Middle East"            2002         0
"Middle East"            2003 .59221643
"Middle East"            2004         0
"Middle East"            2005  5.858264
"Middle East"            2006  9.175097
"Middle East"            2007  7.451975
"Middle East"            2008  7.945778
"Middle East"            2009  6.374172
"Middle East"            2010  6.231583
"Middle East"            2011 4.4821134
"Middle East"            2012  4.335175
"Middle East"            2013  7.054042
"Middle East"            2014  6.882815
"Middle East"            2015  7.979837
"Middle East"            2016   8.19168
"Middle East"            2017  7.772645
"Middle East"            2018  8.790713
"Middle East"            2019  8.470904
"Middle East"            2020  7.209192
"Oceania"                 2002         0
"Oceania"                 2003 2.0340345
"Oceania"                 2004 2.4081006
"Oceania"                 2005 1.0007294
"Oceania"                 2006  .8274303
end

set scheme s1color 

* myaxis is from the Stata Journal 
myaxis order=regions, sort(mean percent) descending 

local xopts xla(2002(3)2020) xtitle("")

local xopts2 xla(2002(3)2020, format(%tyYY)) xtitle("")

separate percent, by(order) veryshortlabel 

twoway connected percent? year , `xopts'

* ,multiline is from SSC 
multiline percent? year , missing `xopts' name(percent1, replace)

* fabplot is from the Stata Journal 
fabplot line percent year, by(order) front(connected) `xopts2' name(percent2, replace)

* tabplot is from the Stata Journal 
tabplot order year [iw=percent], showval(format(%1.0f)) xasis `xopts' ytitle("") separate(regions) barw(0.9) name(percent3, replace)

Graphs below from multiline fabplot tabplot

[Nihat Mugurtay]

Dear Ken Chui and Nick Cox

Your solutions really helped me a lot. But, I want to ask another question Nick Cox,

Nick's "fabplot" is striking for me, because it simplifies complex graphics. I need a similar solution for a wide-format data as below. How can I apply fabplot solution for wide-format? When I run my commands, I got a lot of syntax errors. I want to have a fabplot for each country's percent of aid to Africa.

* Example generated by -dataex-. For more info, type help dataex
clear
input float(time germantperafrica ukafricaper swedishaffper franceaffper euafricaper neterperafrica)
2000         .         .  32.12938         . 32.651512 26.813385
2001         .         .  29.17677         .   39.3174  38.33671
2002  26.18772  33.93139  32.18399 74.357155  40.82576  39.01805
2003  43.22282  44.14436  38.39015  68.13147  45.45539  36.31383
2004   30.3133  48.95535 32.574627  72.79045  44.46368  45.87014
2005 34.267654  51.73958 35.375908  65.55617  45.14848  38.62457
2006  45.59299  67.30754  31.07698  65.67448   43.0117   31.3475
2007 27.988686 38.073322 34.129158  57.85427  45.76632  36.10642
2008 31.785494 35.313072  32.65278  52.28578  44.73254 29.158697
2009 27.482796  38.29399 30.391184   57.7574  43.09824  25.35075
2010  23.33922  38.32915 29.852076  51.23574  44.09679  28.28126
2011 27.828405  39.99601 37.014477  55.40661 35.416862   22.5777
2012  31.49708  41.44385 33.027645  53.42841  41.58309  19.27589
2013 26.863773 37.949303 29.798025  49.31869 37.915318  23.52936
2014 23.470797  39.49032 25.531466  46.02666  41.10809 18.885952
2015  21.42753 36.067783 18.085627   45.5584   38.2148 15.265451
2016 18.450304  33.38219  24.40148   43.3926  37.59255 20.986845
2017 18.907501  33.83401 27.000746  40.39903  39.40042 20.265574
2018 21.965807 30.850166  30.97862  40.17007  39.81797  22.13328
2019 24.717125  28.95965 32.607727  41.28355  45.80464  21.72033
2020 25.549534 27.122955  32.98334  40.07142  44.10706 20.277905
end

[Nick Cox]

fabplot essentially assumes standard Stata conventions here, that you a long layout for panel data, with a single grouping variable Otherwise, the by() option does. not apply.

You need to reshape long but before that you'd need to clean up your variable names.

Here is the renaming done crudely.

* Example generated by -dataex-. For more info, type help dataex
clear
input float(time germantperafrica ukafricaper swedishaffper franceaffper euafricaper neterperafrica)
2000         .         .  32.12938         . 32.651512 26.813385
2001         .         .  29.17677         .   39.3174  38.33671
2002  26.18772  33.93139  32.18399 74.357155  40.82576  39.01805
2003  43.22282  44.14436  38.39015  68.13147  45.45539  36.31383
2004   30.3133  48.95535 32.574627  72.79045  44.46368  45.87014
2005 34.267654  51.73958 35.375908  65.55617  45.14848  38.62457
2006  45.59299  67.30754  31.07698  65.67448   43.0117   31.3475
2007 27.988686 38.073322 34.129158  57.85427  45.76632  36.10642
2008 31.785494 35.313072  32.65278  52.28578  44.73254 29.158697
2009 27.482796  38.29399 30.391184   57.7574  43.09824  25.35075
2010  23.33922  38.32915 29.852076  51.23574  44.09679  28.28126
2011 27.828405  39.99601 37.014477  55.40661 35.416862   22.5777
2012  31.49708  41.44385 33.027645  53.42841  41.58309  19.27589
2013 26.863773 37.949303 29.798025  49.31869 37.915318  23.52936
2014 23.470797  39.49032 25.531466  46.02666  41.10809 18.885952
2015  21.42753 36.067783 18.085627   45.5584   38.2148 15.265451
2016 18.450304  33.38219  24.40148   43.3926  37.59255 20.986845
2017 18.907501  33.83401 27.000746  40.39903  39.40042 20.265574
2018 21.965807 30.850166  30.97862  40.17007  39.81797  22.13328
2019 24.717125  28.95965 32.607727  41.28355  45.80464  21.72033
2020 25.549534 27.122955  32.98334  40.07142  44.10706 20.277905
end

ds time, not 
rename (`r(varlist)') (A B C D E F) 
rename (?) (per=)
reshape long per, i(time) j(whatever) string 
myaxis order=whatever, sort(mean per) descending 
fabplot line per time, by(order) front(connected)

[Nihat Mugurtay]

THANK YOU so much Nick,
Best,

[Nihat Mugurtay]

Dear Nick Cox

I have a dataset below. Now, I am trying to use fabplot as you showed. However, because this is panel data, there are repeated regions for each country. Therefore, in order to use fabplot, I need to group regions and years, and get the sum of "euodatotal" next to the id-year unit. How can I use fabplot in a time-series data with repeated regions? Can you help me to handle this?

Best,

* Example generated by -dataex-. For more info, type help dataex
clear
input str60 recipient float(iso year) str3 ccode float euodatotal str8 region
"Afghanistan"  4 2000 "AFG"  30.01 "Asia"  
"Afghanistan"  4 2001 "AFG"  77.83 "Asia"  
"Afghanistan"  4 2002 "AFG" 226.56 "Asia"  
"Afghanistan"  4 2003 "AFG" 267.59 "Asia"  
"Afghanistan"  4 2004 "AFG" 243.26 "Asia"  
"Afghanistan"  4 2005 "AFG" 288.98 "Asia"  
"Afghanistan"  4 2006 "AFG"  241.5 "Asia"  
"Afghanistan"  4 2007 "AFG"  300.9 "Asia"  
"Afghanistan"  4 2008 "AFG" 318.05 "Asia"  
"Afghanistan"  4 2009 "AFG" 369.53 "Asia"  
"Afghanistan"  4 2010 "AFG"  278.3 "Asia"  
"Afghanistan"  4 2011 "AFG"  334.5 "Asia"  
"Afghanistan"  4 2012 "AFG" 252.24 "Asia"  
"Afghanistan"  4 2013 "AFG" 235.15 "Asia"  
"Afghanistan"  4 2014 "AFG" 334.45 "Asia"  
"Afghanistan"  4 2015 "AFG" 211.39 "Asia"  
"Afghanistan"  4 2016 "AFG" 436.96 "Asia"  
"Afghanistan"  4 2017 "AFG" 439.93 "Asia"  
"Afghanistan"  4 2018 "AFG" 459.73 "Asia"  
"Afghanistan"  4 2019 "AFG" 429.81 "Asia"  
"Afghanistan"  4 2020 "AFG" 368.86 "Asia"  
"Albania"      8 2000 "ALB" 147.37 "Europe"
"Albania"      8 2001 "ALB" 109.78 "Europe"
"Albania"      8 2002 "ALB"  48.27 "Europe"
"Albania"      8 2003 "ALB"  43.62 "Europe"
"Albania"      8 2004 "ALB"  50.75 "Europe"
"Albania"      8 2005 "ALB"  87.56 "Europe"
"Albania"      8 2006 "ALB"  78.06 "Europe"
"Albania"      8 2007 "ALB"  49.24 "Europe"
"Albania"      8 2008 "ALB"  54.99 "Europe"
"Albania"      8 2009 "ALB"  64.75 "Europe"
"Albania"      8 2010 "ALB"  73.18 "Europe"
"Albania"      8 2011 "ALB" 117.73 "Europe"
"Albania"      8 2012 "ALB" 125.39 "Europe"
"Albania"      8 2013 "ALB"  91.29 "Europe"
"Albania"      8 2014 "ALB" 106.78 "Europe"
"Albania"      8 2015 "ALB"  114.3 "Europe"
"Albania"      8 2016 "ALB"  97.88 "Europe"
"Albania"      8 2017 "ALB"  79.29 "Europe"
"Albania"      8 2018 "ALB"  212.9 "Europe"
"Albania"      8 2019 "ALB"  74.45 "Europe"
"Albania"      8 2020 "ALB" 124.34 "Europe"
"Algeria"     12 2000 "DZA"  88.92 "Africa"
"Algeria"     12 2001 "DZA" 166.17 "Africa"
"Algeria"     12 2002 "DZA"  79.29 "Africa"
"Algeria"     12 2003 "DZA"  70.98 "Africa"
"Algeria"     12 2004 "DZA"  82.51 "Africa"
"Algeria"     12 2005 "DZA"  64.13 "Africa"
"Algeria"     12 2006 "DZA" -15.05 "Africa"
"Algeria"     12 2007 "DZA"  83.88 "Africa"
"Algeria"     12 2008 "DZA"  80.23 "Africa"
"Algeria"     12 2009 "DZA"  77.34 "Africa"
"Algeria"     12 2010 "DZA"  50.53 "Africa"
"Algeria"     12 2011 "DZA"  63.09 "Africa"
"Algeria"     12 2012 "DZA"  44.33 "Africa"
"Algeria"     12 2013 "DZA"  48.51 "Africa"
"Algeria"     12 2014 "DZA"  37.09 "Africa"
"Algeria"     12 2015 "DZA"  33.17 "Africa"
"Algeria"     12 2016 "DZA"  61.78 "Africa"
"Algeria"     12 2017 "DZA"  70.27 "Africa"
"Algeria"     12 2018 "DZA"  50.36 "Africa"
"Algeria"     12 2019 "DZA"  53.79 "Africa"
"Algeria"     12 2020 "DZA"  85.46 "Africa"
"Andorra"     20 2000 "ADO"      . "Europe"
"Andorra"     20 2001 "ADO"      . "Europe"
"Andorra"     20 2002 "ADO"      . "Europe"
"Andorra"     20 2003 "ADO"      . "Europe"
"Andorra"     20 2004 "ADO"      . "Europe"
"Andorra"     20 2005 "ADO"      . "Europe"
"Andorra"     20 2006 "ADO"      . "Europe"
"Andorra"     20 2007 "ADO"      . "Europe"
"Andorra"     20 2008 "ADO"      . "Europe"
"Andorra"     20 2009 "ADO"      . "Europe"
"Andorra"     20 2010 "ADO"      . "Europe"
"Andorra"     20 2011 "ADO"      . "Europe"
"Andorra"     20 2012 "ADO"      . "Europe"
"Andorra"     20 2013 "ADO"      . "Europe"
"Andorra"     20 2014 "ADO"      . "Europe"
"Andorra"     20 2015 "ADO"      . "Europe"
"Andorra"     20 2016 "ADO"      . "Europe"
"Andorra"     20 2017 "ADO"      . "Europe"
"Andorra"     20 2018 "ADO"      . "Europe"
"Andorra"     20 2019 "ADO"      . "Europe"
"Andorra"     20 2020 "ADO"      . "Europe"
"Angola"      24 2000 "AGO"  62.36 "Africa"
"Angola"      24 2001 "AGO" 102.76 "Africa"
"Angola"      24 2002 "AGO"  98.33 "Africa"
"Angola"      24 2003 "AGO"  86.99 "Africa"
"Angola"      24 2004 "AGO"  87.99 "Africa"
"Angola"      24 2005 "AGO"  79.15 "Africa"
"Angola"      24 2006 "AGO"  53.11 "Africa"
"Angola"      24 2007 "AGO"  63.48 "Africa"
"Angola"      24 2008 "AGO"  44.97 "Africa"
"Angola"      24 2009 "AGO"  36.36 "Africa"
"Angola"      24 2010 "AGO"  23.81 "Africa"
"Angola"      24 2011 "AGO"  25.86 "Africa"
"Angola"      24 2012 "AGO"  26.06 "Africa"
"Angola"      24 2013 "AGO"  31.17 "Africa"
"Angola"      24 2014 "AGO"  57.89 "Africa"
"Angola"      24 2015 "AGO"  29.58 "Africa"
end

Nihat.

[Nick Cox]

If you have 200 or so countries, nothing will work well with fabplot except after a collapse of some kind by region and year.

[Nihat Mugurtay]

Thank you Nick,

I will check.

Best

[Nihat Mugurtay]

Dear Nick Cox,

Thank you for offering alternative ways of connected plot as you show #3. Actually, fabplot helped me a lot to deal with complex connected plots.
I have another specific question for fabplot.

I tried to create a scatter plot, using the fabplot schema as below. I want to associate "qataraidinteractions" and "qatarbilateral" using a fabplot for different regions (instead of using year). Each region might show a different correlation between "qataraidinteractions" and "qatarbilateral", and I want to visualize this variation.Actually, I am trying to do it without combining many graphs. Fabplot -perhaps- might be a good option, but the second line throws an error.
How can I overcome the problem?

Thank you so much,

Best,

collapse (sum) qataraidinteractions qatarbilateral , by(region year)
myaxis order=region, sort(sum qataraidinteractions qatarbilateral) descending
local xopts xla(2014(2)2020) xtitle("")
local xopts2 xla(2014(2)2020, format(%tyYY)) xtitle("")
fabplot line qataraidinteractions qatarbilateral, by(order) front(scatter) `xopts2' name(percent2, replace)

[Nick Cox]

Data example (after collapse) and error message please.

[Nihat Mugurtay]

Dear Nick Cox,

Thank you for response.

Here, I provide a dataex and other related codes.

* Example generated by -dataex-. For more info, type help dataex
clear
input float year str15 region double(qataraidinteractions qatarbilateral)
2000 "Africa"    0   0
2001 "Africa"    0   0
2002 "Africa"    0   0
2003 "Africa"    0   0
2004 "Africa"    0   0
2005 "Africa"    0   0
2006 "Africa"    0   0
2007 "Africa"    0   0
2008 "Africa"    0   0
2009 "Africa"    0   0
2010 "Africa"    0   0
2011 "Africa"    0   0
2012 "Africa"    0   0
2013 "Africa"    0   0
2014 "Africa"   29  71
2015 "Africa"   34 149
2016 "Africa"   46 105
2017 "Africa"   59 157
2018 "Africa"   60 115
2019 "Africa"   43 137
2020 "Africa"   53  77
2000 "Americas"  0   0
2001 "Americas"  0   0
2002 "Americas"  0   0
2003 "Americas"  0   0
2004 "Americas"  0   0
2005 "Americas"  0   0
2006 "Americas"  0   0
2007 "Americas"  0   0
2008 "Americas"  0   0
2009 "Americas"  0   0
2010 "Americas"  0   0
2011 "Americas"  0   0
2012 "Americas"  0   0
2013 "Americas"  0   0
2014 "Americas"  0  28
2015 "Americas"  0  37
2016 "Americas"  1  38
2017 "Americas" 11  54
2018 "Americas"  5  71
2019 "Americas"  3  74
2020 "Americas" 21  56
2000 "Asia"      0   0
2001 "Asia"      0   0
2002 "Asia"      0   0
2003 "Asia"      0   0
2004 "Asia"      0   0
2005 "Asia"      0   0
2006 "Asia"      0   0
2007 "Asia"      0   0
2008 "Asia"      0   0
2009 "Asia"      0   0
2010 "Asia"      0   0
2011 "Asia"      0   0
2012 "Asia"      0   0
2013 "Asia"      0   0
2014 "Asia"     14  39
2015 "Asia"     29 139
2016 "Asia"     21  86
2017 "Asia"     44 159
2018 "Asia"     34 113
2019 "Asia"     40 144
2020 "Asia"     42 118
2000 "Europe"    0   0
2001 "Europe"    0   0
2002 "Europe"    0   0
2003 "Europe"    0   0
2004 "Europe"    0   0
2005 "Europe"    0   0
2006 "Europe"    0   0
2007 "Europe"    0   0
2008 "Europe"    0   0
2009 "Europe"    0   0
2010 "Europe"    0   0
2011 "Europe"    0   0
2012 "Europe"    0   0
2013 "Europe"    0   0
2014 "Europe"    7  28
2015 "Europe"   11  85
2016 "Europe"    7  33
2017 "Europe"   14  78
2018 "Europe"   13  63
2019 "Europe"   18  44
2020 "Europe"   23  50
2000 "Middle"    0   0
2001 "Middle"    0   0
2002 "Middle"    0   0
2003 "Middle"    0   0
2004 "Middle"    0   0
2005 "Middle"    0   0
2006 "Middle"    0   0
2007 "Middle"    0   0
2008 "Middle"    0   0
2009 "Middle"    0   0
2010 "Middle"    0   0
2011 "Middle"    0   0
2012 "Middle"    0   0
2013 "Middle"    0   0
2014 "Middle"   50  24
2015 "Middle"   63  87
end

------------------ copy up to and including the previous line ------------------

Listed 100 out of 126 observations
Use the count() option to list more

Actually, I want to create a scatter plot or fitplot without using the option below. Instead of using the code below, I want to create scatter or fitplots by using Fabplot.
Thank you so much,

twoway (scatter qataraidinteractions qatarbilateral) (lfit qataraidinteractions qatarbilateral) if region=="Asia"
graph save asia.gph
twoway (scatter qataraidinteractions qatarbilateral) (lfit qataraidinteractions qatarbilateral) if region=="Europe"
graph save europe.gph
twoway (scatter qataraidinteractions qatarbilateral) (lfit qataraidinteractions qatarbilateral) if region=="Africa"
graph save africca.gph
twoway (scatter qataraidinteractions qatarbilateral) (lfit qataraidinteractions qatarbilateral) if region=="Americas"
graph save americas.gph
graph combine asia.gph europe.gph africca.gph americas.gph

The error message is below for my axis. Meanwhile my sample covers the years between 2014 and 2020. I will delete other years later.

Best

myaxis order=region, sort(sum qataraidinteractions qatarbilateral ) descending
sort() invalid; see myaxis
r(198);

[Nick Cox]

Why not give all 126 observations using the count() option of dataex?

[Nihat Mugurtay]

Here, I provide the truncated data. It includes all regions and years.

* Example generated by -dataex-. For more info, type help dataex
clear
input float year str15 region double(qataraidinteractions qatarbilateral)
2014 "Africa"   29  71
2015 "Africa"   34 149
2016 "Africa"   46 105
2017 "Africa"   59 157
2018 "Africa"   60 115
2019 "Africa"   43 137
2020 "Africa"   53  77
2014 "Americas"  0  28
2015 "Americas"  0  37
2016 "Americas"  1  38
2017 "Americas" 11  54
2018 "Americas"  5  71
2019 "Americas"  3  74
2020 "Americas" 21  56
2014 "Asia"     14  39
2015 "Asia"     29 139
2016 "Asia"     21  86
2017 "Asia"     44 159
2018 "Asia"     34 113
2019 "Asia"     40 144
2020 "Asia"     42 118
2014 "Europe"    7  28
2015 "Europe"   11  85
2016 "Europe"    7  33
2017 "Europe"   14  78
2018 "Europe"   13  63
2019 "Europe"   18  44
2020 "Europe"   23  50
2014 "Middle"   50  24
2015 "Middle"   63  87
2016 "Middle"   52  78
2017 "Middle"   51  85
2018 "Middle"   59  75
2019 "Middle"   64  60
2020 "Middle"   84 106
2014 "Ocenia"    0   0
2015 "Ocenia"    0   1
2016 "Ocenia"    0   0
2017 "Ocenia"    0   3
2018 "Ocenia"    0   3
2019 "Ocenia"    1   1
2020 "Ocenia"    1   0
end

------------------ copy up to and including the previous line ------------------

Listed 42 out of 42 observations

[Nick Cox]

Thanks for the data example.

In #10 wanting to insist on xla(2014(2)2020) makes no sense as the x variable is qatarbilateral. -- but the error is that sort() in this call

 
 myaxis order=region, sort(sum qataraidinteractions qatarbilateral) 

doesn't match the specified syntax, which is either

sort(varname)

or

sort(egen_function varname)


fabplot doesn't extend to simultaneous scatter and lfit.

Here is some technique. Again, I corrected the typo pointed out in #3.

* Example generated by -dataex-. For more info, type help dataex
clear
input float year str15 region double(qataraidinteractions qatarbilateral)
2014 "Africa"   29  71
2015 "Africa"   34 149
2016 "Africa"   46 105
2017 "Africa"   59 157
2018 "Africa"   60 115
2019 "Africa"   43 137
2020 "Africa"   53  77
2014 "Americas"  0  28
2015 "Americas"  0  37
2016 "Americas"  1  38
2017 "Americas" 11  54
2018 "Americas"  5  71
2019 "Americas"  3  74
2020 "Americas" 21  56
2014 "Asia"     14  39
2015 "Asia"     29 139
2016 "Asia"     21  86
2017 "Asia"     44 159
2018 "Asia"     34 113
2019 "Asia"     40 144
2020 "Asia"     42 118
2014 "Europe"    7  28
2015 "Europe"   11  85
2016 "Europe"    7  33
2017 "Europe"   14  78
2018 "Europe"   13  63
2019 "Europe"   18  44
2020 "Europe"   23  50
2014 "Middle"   50  24
2015 "Middle"   63  87
2016 "Middle"   52  78
2017 "Middle"   51  85
2018 "Middle"   59  75
2019 "Middle"   64  60
2020 "Middle"   84 106
2014 "Ocenia"    0   0
2015 "Ocenia"    0   1
2016 "Ocenia"    0   0
2017 "Ocenia"    0   3
2018 "Ocenia"    0   3
2019 "Ocenia"    1   1
2020 "Ocenia"    1   0
end

replace region = "Oceania" if region == "Ocenia"
gen qatarsum = qataraid + qatarbil
myaxis order=region, sort(mean qatarsum) descending 
fabplot scatter qataraid qatarbil, by(order) frontopts(ms(Oh) msize(large)) name(G1)

fabplot lfit qataraid qatarbil, by(order) name(G2)