Adding Percent Change to Bar Graph

[Yosef Cohen]

graph bar (sum) cmr_stress1 ccta1 pet_stress1, over(year) ytitle(Patients) title (Cardiac Imaging Volume Data)

Hi,
I am trying to add percent change in addition to "sum" to this graph so that I can see percent increase or decrease every "year" in each of my variables on top of each bar.
Thank you for your help.

[Hemanshu Kumar]

I don't believe you can do this using graph bar (unless you manually edit things with the Graph Editor). But you should be able to do it if you create a bar graph using twoway bar instead (you would then be able to overlay it with a scatter, with invisible marker symbols and labels taken from another variable that tracks the percentage changes you mention).

[Yosef Cohen]

I am not sure what you mean. Would you be able to provide code as an example? Thanks

[Hemanshu Kumar]

Yosef Cohen sure, I would be happy to help. But you'll need to get the ball rolling by giving us a sample of your data using the -dataex- command, as suggested in the Statalist FAQ.

[Nick Cox]

#3 is the wrong way round. Please provide us with a data example! See https://www.statalist.org/forums/help#stata

It just needs to be realistic, and in fact the results of

collapse (sum) cmr_stress1 ccta1 pet_stress1, by(year)
dataex

would be sufficient to give flavour.

[Yosef Cohen]

	
	
* Example generated by -dataex-. To install: ssc install dataex
clear
input float year double(cmr_stress1 ccta1 pet_stress1)
2010   0 22648             39028
2011  77 23859             47850
2012 292 20394             49556
2013 126 18815             69546
2014 346 20575             85451
2015 230 22131             80784
2016   0 25784 86372.09997558594
2017  39 33632             95179
2018   0 41894            108637
2019 327 53357            123049
2020  64 41768            108846
2021 733 62461            118473
end
------------------ copy up to and including the previous line ------------------

Listed 12 out of 12 observations

Thanks

[Nick Cox]

Thanks much for your example. It is hard to know what your graph so far looks like -- but the very different magnitudes are a challenge. With some mixture of brute force and trickery I get this. Next time round I would probably round the % change more.

* Example generated by -dataex- 
clear
input float year double(cmr_stress1 ccta1 pet_stress1)
2010   0 22648             39028
2011  77 23859             47850
2012 292 20394             49556
2013 126 18815             69546
2014 346 20575             85451
2015 230 22131             80784
2016   0 25784 86372.09997558594
2017  39 33632             95179
2018   0 41894            108637
2019 327 53357            123049
2020  64 41768            108846
2021 733 62461            118473
end

rename (cmr_stress1 ccta1 pet_stress1) (q1 q2 q3)

reshape long q, i(year) j(which) 

label def which 1 "better" 2 "labels" 3 "needed"
label val which which 

bysort which (year) : gen toshow = strofreal(100 * (q - q[_n-1])/q[_n-1], "%2.1f") + "%" if _n > 1 & q[_n-1] > 0 

set scheme s1color 

gen nudge = cond(which == 1, 1000, cond(which == 2, 80000, 140000))  

twoway bar q year, base(0) barw(0.8) lcolor(blue) fcolor(blue*0.3) by(which, col(1) legend(off) note("")  yrescale) ytitle("value and % change") subtitle(, fcolor(green*0.1)) xtitle("") || scatter q year, ms(none) mla(toshow) mlabpos(12) mlabcolor(black) xla(2010/2021, tlc(bg)) yla(, ang(h)) || scatter nudge year, ms(none)

[Yosef Cohen]

I am a little confused by your code. What do these two lines do?

 
 bysort which (year) : gen toshow = strofreal(100 * (q - q[_n-1])/q[_n-1], "%2.1f") + "%" if _n > 1 & q[_n-1] > 0   set scheme s1color   gen nudge = cond(which == 1, 1000, cond(which == 2, 80000, 140000))

[Nick Cox]

That's three lines

bysort which (year) : gen toshow = strofreal(100 * (q - q[_n-1])/q[_n-1], "%2.1f") + "%" if _n > 1 & q[_n-1] > 0

That calculates % change. As I wanted to (show how to) show percent signs. I needed to make it a string with trailing "%", I also round to 1 decimal place. Percent change can't be calculated for the first value OR if the previous value was zero.

set scheme s1color

I think you can work this one out.

gen nudge = cond(which == 1, 1000, cond(which == 2, 80000, 140000))

This is a trick. As otherwise some of the bar labels would be cut off by the top axis, I needed to raise the roof of the house. But here there are three houses and a need to raise three roofs. So, I defined high values separately for each panel. In the event I plot this variable with no symbol, so you don't see it, but that's deliberate I just want the side-effect of extending the range on the y axis.

[Yosef Cohen]

Thank you very much for all your time and help. When I tried the code with 5 variables instead of 3 I get this error:

. reshape long q, i(year) j(which)
(note: j = 1 2 3 4 5)
variable id does not uniquely identify the observations
    Your data are currently wide.  You are performing a reshape long.  You specified i(year) and j(which).  In the current wide form, variable year should
    uniquely identify the observations.  Remember this picture:

         long                                wide
        +---------------+                   +------------------+
        | i   j   a   b |                   | i   a1 a2  b1 b2 |
        |---------------| <--- reshape ---> |------------------|
        | 1   1   1   2 |                   | 1   1   3   2  4 |
        | 1   2   3   4 |                   | 2   5   7   6  8 |
        | 2   1   5   6 |                   +------------------+
        | 2   2   7   8 |
        +---------------+
    Type reshape error for a list of the problem observations.
r(9);

. reshape error
(note: j = 1 2 3 4 5)

i (year) indicates the top-level grouping such as subject id.

The data are currently in the wide form; there should be a single
observation per i.

862581 of 862581 observations have duplicate i values:

        +------+
        | year |
        |------|
     1. | 2010 |
     2. | 2010 |
     3. | 2010 |
     4. | 2010 |
     5. | 2010 |
        |------|
     6. | 2010 |
     7. | 2010 |
     8. | 2010 |
     9. | 2010 |
    10. | 2010 |
        |------|
    11. | 2010 |
    12. | 2010 |
    13. | 2010 |
    14. | 2010 |
    15. | 2010 |
        |------|
    16. | 2010 |
    17. | 2010 |
    18. | 2010 |
    19. | 2010 |
    20. | 2010 |
        |------|
    21. | 2010 |
    22. | 2010 |
    23. | 2010 |
    24. | 2010 |
    25. | 2010 |
        |------|
    26. | 2010 |
    27. | 2010 |
    28. | 2010 |
    29. | 2010 |
    30. | 2010 |
        |------|
    31. | 2010 |
    32. | 2010 |
    33. | 2010 |
    34. | 2010 |
    35. | 2010 |
        |------|
    36. | 2010 |
    37. | 2010 |
    38. | 2010 |
    39. | 2010 |
    40. | 2010 |
        |------|
    41. | 2010 |
    42. | 2010 |
    43. | 2010 |
    44. | 2010 |
    45. | 2010 |
        |------|
    46. | 2010 |
    47. | 2010 |
    48. | 2010 |
    49. | 2010 |
    50. | 2010 |
        |------|
    51. | 2010 |
    52. | 2010 |
    53. | 2010 |
    54. | 2010 |
    55. | 2010 |
        |------|
    56. | 2010 |
    57. | 2010 |
    58. | 2010 |
    59. | 2010 |
    60. | 2010 |
        |------|
    61. | 2010 |
    62. | 2010 |
    63. | 2010 |
    64. | 2010 |
    65. | 2010 |
        |------|
    66. | 2010 |
    67. | 2010 |
    68. | 2010 |
    69. | 2010 |
    70. | 2010 |
        |------|
    71. | 2010 |
    72. | 2010 |
    73. | 2010 |
    74. | 2010 |
    75. | 2010 |
        |------|
    76. | 2010 |
    77. | 2010 |
    78. | 2010 |
    79. | 2010 |
    80. | 2010 |
        |------|
    81. | 2010 |
    82. | 2010 |
    83. | 2010 |
    84. | 2010 |
    85. | 2010 |
        |------|
    86. | 2010 |
    87. | 2010 |
    88. | 2010 |
    89. | 2010 |
    90. | 2010 |
        |------|
    91. | 2010 |
    92. | 2010 |
    93. | 2010 |
    94. | 2010 |
    95. | 2010 |
        |------|
    96. | 2010 |
    97. | 2010 |
    98. | 2010 |
    99. | 2010 |
   100. | 2010 |
        |------|
   101. | 2010 |
   102. | 2010 |
   103. | 2010 |
   104. | 2010 |
   105. | 2010 |
        |------|
   106. | 2010 |
   107. | 2010 |
   108. | 2010 |
   109. | 2010 |
   110. | 2010 |
        |------|
   111. | 2010 |
   112. | 2010 |
   113. | 2010 |
   114. | 2010 |
   115. | 2010 |
        |------|
   116. | 2010 |
   117. | 2010 |
   118. | 2010 |
   119. | 2010 |
   120. | 2010 |
        |------|
   121. | 2010 |
   122. | 2010 |
   123. | 2010 |
   124. | 2010 |
   125. | 2010 |
        |------|
   126. | 2010 |
   127. | 2010 |
   128. | 2010 |
   129. | 2010 |
   130. | 2010 |
        |------|
   131. | 2010 |
   132. | 2010 |
   133. | 2010 |
   134. | 2010 |
   135. | 2010 |
        |------|
   136. | 2010 |
   137. | 2010 |
   138. | 2010 |
   139. | 2010 |
   140. | 2010 |
        |------|
   141. | 2010 |
   142. | 2010 |
   143. | 2010 |
   144. | 2010 |
   145. | 2010 |
        |------|
   146. | 2010 |
   147. | 2010 |
   148. | 2010 |
   149. | 2010 |
   150. | 2010 |
        |------|
   151. | 2010 |
   152. | 2010 |
   153. | 2010 |
   154. | 2010 |
   155. | 2010 |
        |------|
   156. | 2010 |
   157. | 2010 |
   158. | 2010 |
   159. | 2010 |
   160. | 2010 |
        |------|
   161. | 2010 |
   162. | 2010 |
   163. | 2010 |
   164. | 2010 |
   165. | 2010 |
        |------|
   166. | 2010 |
   167. | 2010 |
   168. | 2010 |
   169. | 2010 |
   170. | 2010 |
        |------|
   171. | 2010 |
   172. | 2010 |
   173. | 2010 |
   174. | 2010 |
   175. | 2010 |
        |------|
   176. | 2010 |
   177. | 2010 |
   178. | 2010 |
   179. | 2010 |
   180. | 2010 |
        |------|
   181. | 2010 |
   182. | 2010 |
   183. | 2010 |
   184. | 2010 |
   185. | 2010 |
        |------|
   186. | 2010 |
   187. | 2010 |
   188. | 2010 |
   189. | 2010 |
   190. | 2010 |
        |------|
   191. | 2010 |
   192. | 2010 |
   193. | 2010 |
   194. | 2010 |
   195. | 2010 |
        |------|
   196. | 2010 |
   197. | 2010 |
   198. | 2010 |
   199. | 2010 |
   200. | 2010 |
        |------|
   201. | 2010 |
   202. | 2010 |
   203. | 2010 |
   204. | 2010 |
   205. | 2010 |
        |------|
   206. | 2010 |
   207. | 2010 |
   208. | 2010 |
   209. | 2010 |
   210. | 2010 |
        |------|
   211. | 2010 |
   212. | 2010 |
   213. | 2010 |
   214. | 2010 |
   215. | 2010 |
        |------|
   216. | 2010 |
   217. | 2010 |
   218. | 2010 |
   219. | 2010 |
   220. | 2010 |
        |------|
   221. | 2010 |
   222. | 2010 |
   223. | 2010 |
   224. | 2010 |
   225. | 2010 |
        |------|
   226. | 2010 |
   227. | 2010 |
   228. | 2010 |
   229. | 2010 |
   230. | 2010 |
        |------|
   231. | 2010 |
   232. | 2010 |
   233. | 2010 |
   234. | 2010 |
   235. | 2010 |
        |------|
   236. | 2010 |
   237. | 2010 |
   238. | 2010 |
   239. | 2010 |
   240. | 2010 |
        |------|
   241. | 2010 |
   242. | 2010 |
   243. | 2010 |
   244. | 2010 |
   245. | 2010 |
        |------|
   246. | 2010 |
   247. | 2010 |
   248. | 2010 |
   249. | 2010 |
   250. | 2010 |
        |------|
   251. | 2010 |
   252. | 2010 |
   253. | 2010 |
   254. | 2010 |
   255. | 2010 |
        |------|
   256. | 2010 |
   257. | 2010 |
   258. | 2010 |
   259. | 2010 |
   260. | 2010 |
        |------|
   261. | 2010 |
   262. | 2010 |
   263. | 2010 |
   264. | 2010 |
   265. | 2010 |
        |------|
   266. | 2010 |
   267. | 2010 |
   268. | 2010 |
   269. | 2010 |
   270. | 2010 |
        |------|
   271. | 2010 |
   272. | 2010 |
   273. | 2010 |
   274. | 2010 |
   275. | 2010 |
        |------|
   276. | 2010 |
   277. | 2010 |
   278. | 2010 |
   279. | 2010 |
   280. | 2010 |
        |------|
   281. | 2010 |
   282. | 2010 |
   283. | 2010 |
   284. | 2010 |
   285. | 2010 |
        |------|
   286. | 2010 |
   287. | 2010 |
   288. | 2010 |
   289. | 2010 |
   290. | 2010 |
        |------|
   291. | 2010 |
   292. | 2010 |
   293. | 2010 |
   294. | 2010 |
   295. | 2010 |
        |------|
   296. | 2010 |
   297. | 2010 |
   298. | 2010 |
   299. | 2010 |
   300. | 2010 |
        |------|
   301. | 2010 |
   302. | 2010 |
   303. | 2010 |
   304. | 2010 |
   305. | 2010 |
        |------|
   306. | 2010 |
   307. | 2010 |
   308. | 2010 |
   309. | 2010 |
   310. | 2010 |
        |------|
   311. | 2010 |
   312. | 2010 |
   313. | 2010 |
   314. | 2010 |
   315. | 2010 |
        |------|
   316. | 2010 |
   317. | 2010 |
   318. | 2010 |
   319. | 2010 |
   320. | 2010 |
        |------|
   321. | 2010 |
   322. | 2010 |
   323. | 2010 |
   324. | 2010 |
   325. | 2010 |
        |------|
   326. | 2010 |
   327. | 2010 |
   328. | 2010 |
   329. | 2010 |
   330. | 2010 |
        |------|
   331. | 2010 |
   332. | 2010 |
   333. | 2010 |
   334. | 2010 |
   335. | 2010 |
        |------|

[Nick Cox]

So, you have other variables which complicate the question mightily. My code in #7 starts from the collapse results. It is not intended to, and will not, work correctly with your full dataset.