Dear All,
I'm trying to produce a coefficient plot using coefplot (or rspike) to organize coefficients and add labels appropriately from 15 regressions looking at 5 different outcome variables and each with 3 different independent variables. My code is the following:
reg y1 x1 controls,r
eststo result1_1
reg y1 x2 controls,r
eststo result1_2
reg y1 x3 controls,r
eststo result1_3
reg y2 x1 controls,r
eststo result2_1
reg y2 x2 controls,r
eststo result2_2
reg y2 x3 controls,r
eststo result2_3
reg y3 x1 controls,r
eststo result3_1
reg y3 x2 controls,r
eststo result3_2
reg y3 x3 controls,r
eststo result3_3
reg y4 x1 controls,r
eststo result4_1
reg y4 x2 controls,r
eststo result4_2
reg y4 x3 controls,r
eststo result4_3
reg y5 x1 controls,r
eststo result5_1
reg y5 x2 controls,r
eststo result5_2
reg y5 x3 controls,r
eststo result5_3
coefplot (result1_1, offset(-0.38) keep(x1)) (result1_2, offset(-0.33) keep(x2)) (result1_3, offset(-0.28) keep(x3)) ///
(result2_1, offset(-0.15) keep(x1)) (result2_2, offset(-0.1) keep(x2)) (result2_3, offset(-0.05) keep(x3)) ///
(result3_1, offset(0.09) keep(x1)) (result3_2, offset(0.135) keep(x2)) (result3_3, offset(-0.09) keep(x3)) ///
(result4_1, offset(0.3) keep(x1)) (result4_2, offset(0.35) keep(x2)) (result4_3, offset(0.4) keep(x3)) ///
(result5_1,offset(0.55) keep(x1)) (result5_2,offset(0.6) keep(x2)) (result5_3,offset(0.65) keep(x3)), ///
vertical ///
xlabel(, notick labgap(0)) legend(off)
The graph I got looks like the following. It seems that coefplot grouped by the x variables instead of by outcome variables, i.e., x1 x1 x1 x1 x1 || x2 x2 x2 x2 x2 || x3 x3 x3 x3 x3 instead of x1 x2 x3 || x1 x2 x3 || x1 x2 x3 || x1 x2 x3 || x1 x2 x3 and I don't know how to label my x-axis appropriately
I can potentially use rspike but I don't know how to group my coefficients without creating subplots. The end results I want is something as the following graph shows. Many thanks!
I'm trying to produce a coefficient plot using coefplot (or rspike) to organize coefficients and add labels appropriately from 15 regressions looking at 5 different outcome variables and each with 3 different independent variables. My code is the following:
reg y1 x1 controls,r
eststo result1_1
reg y1 x2 controls,r
eststo result1_2
reg y1 x3 controls,r
eststo result1_3
reg y2 x1 controls,r
eststo result2_1
reg y2 x2 controls,r
eststo result2_2
reg y2 x3 controls,r
eststo result2_3
reg y3 x1 controls,r
eststo result3_1
reg y3 x2 controls,r
eststo result3_2
reg y3 x3 controls,r
eststo result3_3
reg y4 x1 controls,r
eststo result4_1
reg y4 x2 controls,r
eststo result4_2
reg y4 x3 controls,r
eststo result4_3
reg y5 x1 controls,r
eststo result5_1
reg y5 x2 controls,r
eststo result5_2
reg y5 x3 controls,r
eststo result5_3
coefplot (result1_1, offset(-0.38) keep(x1)) (result1_2, offset(-0.33) keep(x2)) (result1_3, offset(-0.28) keep(x3)) ///
(result2_1, offset(-0.15) keep(x1)) (result2_2, offset(-0.1) keep(x2)) (result2_3, offset(-0.05) keep(x3)) ///
(result3_1, offset(0.09) keep(x1)) (result3_2, offset(0.135) keep(x2)) (result3_3, offset(-0.09) keep(x3)) ///
(result4_1, offset(0.3) keep(x1)) (result4_2, offset(0.35) keep(x2)) (result4_3, offset(0.4) keep(x3)) ///
(result5_1,offset(0.55) keep(x1)) (result5_2,offset(0.6) keep(x2)) (result5_3,offset(0.65) keep(x3)), ///
vertical ///
xlabel(, notick labgap(0)) legend(off)
The graph I got looks like the following. It seems that coefplot grouped by the x variables instead of by outcome variables, i.e., x1 x1 x1 x1 x1 || x2 x2 x2 x2 x2 || x3 x3 x3 x3 x3 instead of x1 x2 x3 || x1 x2 x3 || x1 x2 x3 || x1 x2 x3 || x1 x2 x3 and I don't know how to label my x-axis appropriately
I can potentially use rspike but I don't know how to group my coefficients without creating subplots. The end results I want is something as the following graph shows. Many thanks!
