我想基于频率创建并排条形图。在Side-by-side bars in bar plot找到了一个R例子。这可以在Stata中找到吗?
假设我有这些数据:
lock(&sched.lock)
我可以使用用户编写的命令clear
set obs 1000
set seed 1
gen var1 = runiform()*10
set seed 2
gen var2 = runiform()*10
set seed 3
gen var3 = runiform()*10
gen var1_value = "A" if var1<3
replace var1_value = "B" if var1>=3 & var1<=9
replace var1_value = "C" if var1>9
gen var2_value = "A" if var2<4
replace var2_value = "B" if var2>=4 & var2<=6
replace var2_value = "C" if var2>6
gen var3_value = "A" if var3<1
replace var3_value = "B" if var3>=1 & var3<=3
replace var3_value = "C" if var3>3
和tabplot
来单独绘制这些命令:
catplot
然而,我希望这三个在同一个情节中,并列价值。如果可能的话,我希望每个值并排放置许多图,而不是本例中的3个。
答案 0 :(得分:1)
有几种方法可以做到这一点。最简单的可能是简单的数据重组。按照您的示例,我假设先前为ssc inst catplot
。为不同颜色的razzmatazz添加asyvars
等选项。
clear
set obs 1000
set seed 1
gen var1 = runiform()*10
set seed 2
gen var2 = runiform()*10
set seed 3
gen var3 = runiform()*10
gen var1_value = "A" if var1<3
replace var1_value = "B" if var1>=3 & var1<=9
replace var1_value = "C" if var1>9
gen var2_value = "A" if var2<4
replace var2_value = "B" if var2>=4 & var2<=6
replace var2_value = "C" if var2>6
gen var3_value = "A" if var3<1
replace var3_value = "B" if var3>=1 & var3<=3
replace var3_value = "C" if var3>3
forval j = 1/3 {
clonevar value`j' = var`j'_value
}
gen id = _n
reshape long value, i(id) j(which)
catplot which value, asyvars
catplot value which, asyvars