stata matrix相关命令学习笔记

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

共同学习，如有问题欢迎批评指正。
代码相关原文见帖：
【分享学习】空间面板向量自回归Sp P-var论文数据及代码
https://bbs.pinggu.org/forum.php?mod=viewthread&tid=10906475&from^^uid=11374248


以下为sp pvar代码解读

1. *==============
   
2. * Main Results
   
3. *==============
   
4. 
   
5. ***1. estimate the model and run two stability tests (time and spatial stability)
   
6. 
   
7. pvarstablecivo //time stability
   
8. 
   
9. display("=======================")
   
10. //computation for spatial stability (specific for two endogenous variables and one lag only)
    
11. display("the largest modulus of time-spatial roots is") //spatial and time roots
    
12. 
    
13. scalar bigN=e(N_clust) // N_clust short for Number of clusters  聚类变量的个数；scalar定义一个标量，用以后续计算
    
14. matrix B=e(b) // e(b)是coefficient vector,回归后保存结果系数列向量的矩阵e(b)为B  
    
15. scalar smalln=wordcount(e(depvar)) // wordcount计算内生变量的个数：2
    
16. 
    
17. 
    
18. matrix A1=[B[1,"lights:L.lights".."lights:L.odas"] \ B[1,"odas:L.lights".."odas:L.odas"]] // 定义一个名为A1的2行2列的矩阵，逗号作为列的分隔，反斜线作为行的分隔，注意反斜线的前后要加空格，否则斜线和数字挨到一起stata不识别；选取的是时间滞后项
    
19. 
    
20. matrix colnames A1= lig aid // 定义列名,分别为lig、aid
    
21. matrix rownames A1= lig aid // 定义行名,分别为lig、aid
    
22. matrix coleq A1= " " // matrix coleq: Reset column names and interpret simple names as equation names. matrix roweq and coleq also reset the row and column names of an already existing matrix, but if a simple name (a name without a colon) is specified, it is interpreted as an equation name.在执行这一步之前可以执行：matrix list A1，能够发现矩阵行标题上有多一行带冒号的文字，此命令用于定义冒号前的名称，输入空格或者没有内容则消除带冒号的标题行
    
23. 
    
24. matrix S=[B[1,"lights:lights36".."lights:odas36"] \ B[1,"odas:lights36".."odas:odas36"]] // 选取的是空间滞后项
    
25. matrix colnames S= lig aid
    
26. matrix rownames S= lig aid
    
27. matrix coleq S= " "
    
28. 
    
29. spatwmat using "CC36" , name(CC36) // CC36是The contiguity matrix (either 1 if
    
30. districts i and j are neighbors or 0 if j is not a neighbor of i, with a diagonal
    
31. of zeros by construction)，是空间矩阵，原为dta文件，通过spatwmat 转为矩阵保存
    
32. 
    
33. matrix I_n=I(smalln) // 生成smalln维的单位矩阵（smalln代表一个标量）

复制代码

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

分享0 收藏2 回帖

关键词：matrix Stata 学习笔记 tata Mat

1. ***2. run the autocorrelation and cross-sectional dependence checks
   
2. //these pvars show that the residuals of the (1,1) specification are autocorrelated, while the baseline has cleaner residuals
   
3. 
   
4. qui pvarcivo lights odas if ash==1 | dark==1, lags(1) instlag(1) gmmopts(two  winitial(i) wmatrix(cl id)) vce(cle id) spatial(lights36 odas36) exog(rains) ivspt(rains36) overid sig2 td
   
5. gen res1=res00lights // pvarcivo中的option sig2 produces a series of residuals resid00‘var’ for each of the endogenous variables ‘var’ of the model
   
6. gen res2=res00odas
   
7. pvarcivo res1 res2 if ash==1 | dark==1, lags(1) instlag(1) gmmopts(two  winitial(i) wmatrix(cl id)) vce(cle id) overid
   
8. 
   
9. qui pvarcivo lights odas if ash==1 | dark==1, lags(1) instlag(2/3) gmmopts(two  winitial(i) wmatrix(cl id)) vce(cle id) spatial(lights36 odas36) exog(rains) ivspt(rains36) overid sig2 td
   
10. replace res1= res00lights
    
11. replace res2= res00odas
    
12. pvarcivo res1 res2 if ash==1 | dark==1, lags(1) instlag(1) gmmopts(two  winitial(i) wmatrix(cl id)) vce(cle id) overid
    
13. 
    
14. //estiamte the dynamic panel version of the model with xtabond2; also show relative importance of subsets of IVs; see Section 4.1 of the paper

复制代码