搜索
人大经济论坛 › 附件下载

附件下载

所在主题：
文件名:  Intro to data visualization.do
资料下载链接地址: https://bbs.pinggu.org/a-1905631.html
附件大小:

Intro to data visualization


  2. *******************************************************************************
  3. *******************************************************************************
  4. ***** *****
  5. ***** *****
  6. ***** Intro to data visualization *****
  7. ***** Oscar Torres-Reyna *****
  8. ***** DSS Princeton University *****
  9. ***** *****
  10. ***** *****
  11. ***** *****
  12. *******************************************************************************
  13. *******************************************************************************


  16. * NOTE, commands should be type either in the command window, see here page 5 here
  17. * http://dss.princeton.edu/training/StataTutorial.pdf#page=5
  18. * on in a do-file, see here page 9:
  19. * http://dss.princeton.edu/training/StataTutorial.pdf#page=9

  21. * Stata has a color-coded system, see here page 13
  22. * http://dss.princeton.edu/training/StataTutorial.pdf#page=13
  23. ******* Setting working directory
  24. *NOTE: If using Mac go to File->Change Working Directory, and select the folder

  26. *cd "H:"

  28. ******* Creating a log file.
  29. qui log c
  30. qui log using mylog.log, replace
  31. use "http://www.princeton.edu/~otorres/wdipol.dta", clear

  33. * See the data
  34. browse
  35. ******* Getting to know your data
  36. describe
  37. summarize
复制代码
  1. ******* Line graphs
  2. set scheme s1color
  3. line unemp unempf unempm year if country=="United States"


  6. summarize unemp unempf unempm
  7. replace unemp=. if unemp==0
  8. replace unempf=. if unempf==0
  9. replace unempm=. if unempm==0
  10. summarize unemp unempf unempm


  13. line unemp unempf unempm year if country=="United States"


  16. twoway line unemp unempf unempm year if country=="United States", ///
  17. title("Unemployment rate in the US, 1980-2012") ///
  18. legend(label(1 "Total") label(2 "Females") label(3 "Males")) ///
  19. lpattern(solid dash dot) ///
  20. ytitle("Percentage")

  22. twoway connected unemp unempf unempm year if country=="United States", ///
  23. title("Unemployment rate in the US, 1980-2012") ///
  24. legend(label(1 "Total") label(2 "Females") label(3 "Males")) ///
  25. msymbol(circle diamond square) ///
  26. ytitle("Percentage")


  29. twoway connected unemp year if country=="United States" | ///
  30. country=="United Kingdom" | ///
  31. country=="Australia" | ///
  32. country=="Qatar", ///
  33. by(country, title("Unemployment")) ///
  34. msymbol(circle_hollow)


  37. twoway (connected unemp year if country=="United States", msymbol(dh)) ///
  38. (connected unemp year if country=="United Kingdom", msymbol(th)) ///
  39. (connected unemp year if country=="Australia", msymbol(sh)) ///
  40. (connected unemp year if country=="Qatar", ///
  41. title("Unemployment") ///
  42. msymbol(ch) ///
  43. legend(label(1 "USA") label(2 "UK") label(3 "Australia") label(4 "Qatar")))


  46. twoway connected gdppc year if gdppc>40000, by(country) msymbol(diamond)


  49. bysort year: egen gdppc_mean=mean(gdppc)
  50. bysort year: egen gdppc_median=median(gdppc)
  51. twoway connected gdppc gdppc_mean year if country=="United States" | ///
  52. country=="United Kingdom" | ///
  53. country=="Australia" | ///
  54. country=="Qatar", ///
  55. by(country, title("GDP pc (PPP, 2005=100)")) ///
  56. legend(label(1 "GDP-PC") label(2 "Mean GDP-PC")) ///
  57. msymbol(circle_hollow)


  60. help twoway line
  61. help twoway connected
复制代码















  1. ******* Graph markers




  6. palette symbolpalette



  10. palette linepalette



  14. palette color green


  17. /*
  18. ssc install showmarkers

  20. showmarkers , over(msymbol)

  22. showmarkers , over(mcolor)

  24. showmarkers , over(mlpattern)
  25. */
复制代码



  1. ******* Bar graphs






  8. graph hbar (mean) gdppc /*Mean is the default*/





  14. graph hbar (mean) gdppc, over(country, sort(1) descending)





  20. graph hbar (mean) gdppc, over(country, sort(1) descending label(labsize(*0.5)))





  26. graph hbar (mean) gdppc (median) gdppc if gdppc>40000, ///
  27. over(country, sort(1) descending label(labsize(*1))) ///
  28. legend(label(1 "GDPpc (mean)") label(2 "GDPpc (median)"))




  33. help graph bar
复制代码









  1. ******* Box plots
  2. * Need to recode polity2

  4. recode polity2 (-10/-6=1 "Autocracy") ///
  5. (-5/6=2 "Anocracy") ///
  6. (7/10=3 "Democracy") ///
  7. (else=.), ///
  8. gen(regime) label(polity_rec)

  10. tab regime /* Frequency */
  11. tab regime, nolabel /* See numeric values*/
  12. tab country regime /* Cross tabulations */
  13. tab country regime, row /* Adding percent per row */

  15. help tab

  17. graph hbox gdppc

  19. graph hbox gdppc if gdppc<40000

  21. graph box gdppc, over(regime) yline(4517.94) marker(1,mlabel(country))

  23. help graph box
复制代码



  1. ******* Scatterplots


  4. * scatter y x



  8. scatter import export



  12. #d;
  13. twoway scatter import export || scatter import export if export>1000000,
  14. mlabel(country);
  15. #d cr



  19. twoway (scatter import export, ytitle("Imports") xtitle("Exports")) ///
  20. (scatter import export if export>1000000, mlabel(country) legend(off)) ///
  21. (lfit import export, note("Constant values, 2005, millions US[ DISCUZ_CODE_5 ]quot;))



  25. *bysort year: egen gdppc_mean=mean(gdppc)


  28. twoway (scatter gdppc year, jitter(13)) ///
  29. (connected gdppc_mean year, msymbol(diamond)) , xlabel(1980(1)2012, angle(90))



  33. help twoway scatter


  36. ******* Scatterplot matrix



  40. gr matrix gdppc unemp unempf unempm export import trade polity2, ///
  41. maxis(ylabel(none) xla(none))



  45. gr matrix gdppc unemp unempf unempm export import trade polity2, ///
  46. half maxis(ylabel(none) xla(none))



  50. help graph matrix
复制代码












  1. ******* Histograms




  6. hist gdppc



  10. /* Shows density*/




  15. hist gdppc, frequency




  20. /*Shows frequency*/




  25. hist gdppc, kdensity



  29. /* Combo histogram and density plot */



  33. hist gdppc, kdensity normal





  39. /* Adding a normal curve */




  44. hist gdppc, kdensity normal bin(20)




  49. hist gdppc if country=="United States" | country=="United Kingdom", bin(10) ///
  50. by(country)




  55. twoway hist gdppc if country=="United States", bin(10) || ///
  56. hist gdppc if country=="United Kingdom", bin(10) ///
  57. fcolor(none) lcolor(black) legend(label(1 "USA") label(2 "UK"))




  62. help hist
复制代码












  1. ******* Setup panel data

  3. * See http://dss.princeton.edu/training/Panel101.pdf

  5. *xtset country year
  6. /*Gives an error, 'country' is string*/

  8. encode country, gen(country1)

  10. /*Assign numeric value to strings*/
  11. xtset country1 year


  14. /*No error, 'country1' is coded variable*/

  16. xtline gdppc

  18. xtline gdppc if gdppc>39000, overlay

  20. help xtline

  22. ******* Combining graphs

  24. graph drop _all /*Drop graphics saved in memory*/

  26. hist gdppc if country=="United States", name(gdppc, replace)


  29. line unemp year if country=="United States", name(unemp, replace)


  32. graph combine gdppc unemp, col(1)

  34. help graph combine
  35. ******* Scatterplots with linear fit and confidence intervals
  36. use "http://dss.princeton.edu/training/students.dta", clear

  38. twoway (lfitci sat age) ///
  39. (scatter sat age, mlabel(lastname)),///
  40. title("SAT scores by age") ytitle("Sat")
  41. * Changing position

  43. generate position=3
  44. replace position=6 if lastname=="DOE01"
  45. replace position=6 if lastname=="DOE10"
  46. replace position=12 if lastname=="DOE14"
  47. replace position=12 if lastname=="DOE29"

  49. #d;
  50. twoway (lfitci sat age)
  51. (scatter sat age, mlabel(lastname)mlabv(position)
  52. jitter(21)), title("SAT scores by age") ytitle("Sat");
  53. #d cr

  55. * Without confidence intervals
  56. #d;
  57. twoway (lfit sat age)
  58. (scatter sat age, mlabel(lastname)mlabv(position)
  59. jitter(21)),title("SAT scores by age") ytitle("Sat");
  60. #d cr
  61. help twoway lfit
  62. help twoway lfitci
  63. ******* Plotting categorical variables

  65. ******* Mosaic plots (a.k.a spineplots)


  68. * May need to install it, type:

  70. ssc install spineplot

  72. use "http://dss.princeton.edu/training/students.dta", clear

  74. encode gender, gen(gender1)

  76. /* Assign numeric values to categories in string format*/

  78. encode major, gen(major1)

  80. spineplot gender1 major1

  82. bysort gender1 major1: gen gendermajor = _N

  84. spineplot gender1 major1, text(gendermajor)

  86. spineplot gender1 major1, percent bar1(bcolor(yellow)) ///
  87. bar2(bcolor(green)) text(gendermajor)

  89. * See the graphs here:
  90. * http://www.princeton.edu/~otorres/mosaic1.pdf
  91. * http://www.princeton.edu/~otorres/mosaic2.pdf

  93. ******* Using catplot

  95. * Chernoff faces
  96. * Few cases, each face is a row case.

  98. use "http://www.princeton.edu/~otorres/chernoff.dta", clear

  100. net install gr0038, from(http://www.stata-journal.com/software/sj9-3)

  102. /*User-written command, need to install*/

  104. chernoff, hdark(gdppc) bdens(trade) nose(unemp) mcurv(polity2) ///
  105. order(gdppc) ilabel(country)

  107. * See the graph here:
  108. * http://www.princeton.edu/~otorres/chernoff.pdf





  114. * Do not forget to close the log
  115. log close
复制代码































/*讲义下载*/

  1. copy "http://dss.princeton.edu/training/Visual101.pdf" "Visual101.pdf"
复制代码


    熟悉论坛请点击新手指南
下载说明
1、论坛支持迅雷和网际快车等p2p多线程软件下载，请在上面选择下载通道单击右健下载即可。
2、论坛会定期自动批量更新下载地址,所以请不要浪费时间盗链论坛资源,盗链地址会很快失效。
3、本站为非盈利性质的学术交流网站,鼓励和保护原创作品，拒绝未经版权人许可的上传行为。本站如接到版权人发出的合格侵权通知，将积极的采取必要措施；同时，本站也将在技术手段和能力范围内，履行版权保护的注意义务。
(如有侵权，欢迎举报)
二维码

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

京ICP备16021002号-2 京B2-20170662号 京公网安备 11010802022788号 论坛法律顾问：王进律师 知识产权保护声明   免责及隐私声明

GMT+8, 2025-12-31 21:49