Coefplot - issue with x-axis

[Konstantina Boutsioukou]

Dear all,

I have the following dataset:


* Example generated by -dataex-. To install: ssc install dataex
clear
input double(elite yeard) float(edu_f occup_f) byte sex_stud float nationality double lambda_1
0 0 1 1 2 1 1.0369200891134476
0 0 1 1 1 1 1.2773466215065548
0 0 1 1 2 1 1.0369200891134476
0 1 1 2 2 1 .8312793625071068
0 11 1 2 1 1 .8729843442323654
0 0 1 2 2 1 .7560635291996983
0 4 1 1 2 1 .7926716817333047
0 5 1 1 2 0 .
0 0 1 1 2 1 1.0369200891134476
0 7 1 4 2 1 .7755610866588346
0 10 1 4 2 1 1.0160053143124155
0 8 1 1 2 1 .8887555852577639
0 11 1 2 2 1 .6680324038100115
0 10 1 1 1 1 1.3679320903962948
0 8 1 4 1 1 1.0107202621427354
0 5 1 1 2 1 1.1311930843211504
0 10 1 4 2 1 1.0160643653441264
1 0 1 1 2 1 1.0349573006908248
0 0 1 2 2 1 .7570583682780064
0 6 1 2 1 1 1.0072529596758164
0 12 1 4 1 1 1.367753191436586
0 0 1 2 1 1 .9730738007889017
0 8 1 1 1 1 1.1170189979032021
0 6 1 1 1 1 1.3144360777565052
0 6 1 1 2 1 1.0726927331894347
0 6 1 1 2 0 .
1 11 1 1 2 1 .9341283974570147
0 2 1 2 1 1 .996309555751049
0 9 1 2 1 1 .8333933093421787
0 9 1 1 1 1 1.1242444246785002
0 5 1 3 1 1 1.1102436654585008
0 12 1 1 2 1 1.2316692792947987
0 3 1 2 2 1 .7570828902424422
0 4 1 1 2 1 .7938476228873457
0 2 1 2 2 1 .7785689802872727
0 11 1 2 2 1 .6680324038100115
1 0 1 1 1 1 1.2773466215065548
0 1 1 1 2 0 .
0 9 1 1 2 1 .8935301157318588
0 8 1 1 2 1 .8887555852577639
0 12 1 1 2 1 1.2316692792947987
0 10 1 1 1 1 1.369042565855686
0 1 1 1 2 1 1.1202806903795384
0 10 1 1 1 1 1.369042565855686
0 0 1 1 2 1 1.0369200891134476
0 11 1 1 2 0 1.4906795149829437
0 1 1 2 2 0 1.3693299189081976
0 2 1 1 1 1 1.3026266331568117
0 6 1 1 2 1 1.0726927331894347
0 0 1 3 2 1 .7965379426985457
0 5 1 1 1 1 .
0 5 1 1 2 1 1.1311930843211504
0 6 1 1 2 1 1.071909703671093
0 3 1 1 2 1 1.0355797837686915
0 0 1 2 2 1 .7570583682780064
0 11 1 2 1 1 .8728730582057933
0 9 1 1 2 1 .8935301157318588
0 8 1 1 1 1 1.1155525294884263
0 0 1 1 2 1 1.0369200891134476
0 12 1 1 1 1 1.485016757386524
0 5 1 4 2 1 1.0242147955656187
0 12 1 1 2 1 1.2316692792947987
0 1 1 4 2 1 1.015164129725451
0 2 1 1 1 1 1.3037204311079884
0 9 1 1 1 1 1.124611978392552
0 12 1 1 2 1 1.2314797359971197
0 12 1 1 1 1 1.4848160304977387
1 5 1 3 2 1 .8818733881434547
0 9 1 1 1 1 1.124611978392552
0 8 1 1 1 1 1.1170189979032021
0 3 1 1 1 1 .
0 6 1 1 1 1 1.3148877088987125
0 0 1 1 1 1 1.2773466215065548
0 10 1 1 2 1 1.1225602859771444
0 7 1 1 2 1 .8716736465959619
0 6 1 1 2 0 .
0 10 1 1 1 0 .
1 11 1 1 2 1 .9341283974570147
0 3 1 1 2 1 1.0355797837686915
1 5 1 1 2 0 .
0 6 1 1 2 1 1.0714881580017552
0 9 1 1 2 1 .8935301157318588
0 12 1 1 1 1 1.485016757386524
0 11 1 1 1 1 1.1664940215408328
0 6 1 4 2 1 .9680181749086781
0 2 1 1 1 1 1.3037204311079884
0 4 1 1 1 1 1.0115907523658474
0 8 1 1 1 0 1.7014600969048974
0 8 1 4 1 1 1.0107202621427354
0 2 1 1 1 1 1.3023693110016734
0 1 1 2 2 1 .8312793625071068
0 3 1 1 1 1 1.2764815825186624
0 5 1 1 2 1 1.1311930843211504
0 12 1 3 2 1 .972769328431093
0 10 1 1 2 1 1.1215196595130927
0 9 1 1 1 1 1.122223631833564
0 11 1 2 1 1 .8729843442323654
1 7 1 1 2 1 .8734539182656972
0 2 1 1 2 0 .
0 6 1 1 2 1 1.0717892559347058
end
label values elite elite
label def elite 0 "Non-elite", modify
label def elite 1 "Elite", modify
label values yeard yeard
label def yeard 0 "2004", modify
label def yeard 1 "2005", modify
label def yeard 2 "2006", modify
label def yeard 3 "2007", modify
label def yeard 4 "2008", modify
label def yeard 5 "2009", modify
label def yeard 6 "2010", modify
label def yeard 7 "2011", modify
label def yeard 8 "2012", modify
label def yeard 9 "2013", modify
label def yeard 10 "2014", modify
label def yeard 11 "2015", modify
label def yeard 12 "2016", modify
label values edu_f edu_f
label def edu_f 1 "Primary education or below", modify
label values occup_f occup_f
label def occup_f 1 "Blue collar", modify
label def occup_f 2 "Low skilled white collar", modify
label def occup_f 3 "High skilled white collar", modify
label def occup_f 4 "Unemployed", modify
label values sex_stud sex_stud
label def sex_stud 1 "Male", modify
label def sex_stud 2 "Female", modify
label values nationality nationality
label def nationality 0 "Non-Greek", modify
label def nationality 1 "Greek", modify
[/CODE]


I am using the following code to plot the coefficients of two models:
eststo: reg elite i.yeard i.edu_f#ibn.yeard i.occup_f i.sex_stud i.nationality, baselevels
eststo: reg elite i.yeard i.edu_f#ibn.yeard i.occup_f i.sex_stud i.nationality lambda_1, baselevels

coefplot est1 est2, nooffset vertical yline(0) keep(2.edu_f#*.yeard 3.edu_f#*.yeard 4.edu_f#*.yeard) ///
groups(2.edu_f#?.yeard="{bf:Secondary education}" ///
3.edu_f#?.yeard="{bf:Vocational education}" ///
4.edu_f#?.yeard="{bf:Tertiary education}") ///
graphregion(fcolor(white)) ////
byopts(xrescale) ///
legend(order(2 "Linear Probability Model" 4 "Linear Probability Model with Heckman correction") size(small)) ///
coeflabels(2.edu_f#0.yeard="2004" 2.edu_f#1.yeard="2005" 2.edu_f#2.yeard="2006" 2.edu_f#3.yeard="2007" 2.edu_f#4.yeard="2008" 2.edu_f#5.yeard="2009" 2.edu_f#6.yeard="2010" 2.edu_f#7.yeard="2011" 2.edu_f#8.yeard="2012" 2.edu_f#9.yeard="2013" 2.edu_f#10.yeard="2014" 2.edu_f#11.yeard="2015" 2.edu_f#12.yeard="2016" 3.edu_f#0.yeard="2004" 3.edu_f#1.yeard="2005" 3.edu_f#2.yeard="2006" 3.edu_f#3.yeard="2007" 3.edu_f#4.yeard="2008" 3.edu_f#5.yeard="2009" 3.edu_f#6.yeard="2010" 3.edu_f#7.yeard="2011" 3.edu_f#8.yeard="2012" 3.edu_f#9.yeard="2013" 3.edu_f#10.yeard="2014" 3.edu_f#11.yeard="2015" 3.edu_f#12.yeard="2016" 4.edu_f#0.yeard="2004" 4.edu_f#1.yeard="2005" 4.edu_f#2.yeard="2006" 4.edu_f#3.yeard="2007" 4.edu_f#4.yeard="2008" 4.edu_f#5.yeard="2009" 4.edu_f#6.yeard="2010" 4.edu_f#7.yeard="2011" 4.edu_f#8.yeard="2012" 4.edu_f#9.yeard="2013" 4.edu_f#10.yeard="2014" 4.edu_f#11.yeard="2015" 4.edu_f#12.yeard="2016", notick labgap(2) angle(vertical))
graph export "$coefplots/edu_father_Heck_2.png", replace

The code produces the following graph:

However, I do not understand why there is a bigger gap between 2013 and 2014 across all educational categories.

Any help would be greatly appreciated!

Best,
Konstantina

[Andrew Musau]

Should be due to the fact that yeard is coded 0 to 15, and you jump from 1 digit to 2 digits between 9=2013 and 10=2014. Indicators require integer values, so just use the raw year values which are all 4 digits.

lab drop yeard
replace yeard= yeard+2004
eststo: reg elite i.yeard i.edu_f#ibn.yeard i.occup_f i.sex_stud i.nationality, baselevels
eststo: reg elite i.yeard i.edu_f#ibn.yeard i.occup_f i.sex_stud i.nationality lambda_1, baselevels

[Konstantina Boutsioukou]

Dear Andrew,

Thank you very much for your prompt reply. I fixed the issue in the graph now.

Best,
Konstantina