指标的正负号与过往文献有误

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

因论文要求，需要建构股价信息效率指标
参照外文，以Hou的方式建构
国内也有很多中文期刊学者加以引用这个做法建构股价信息效率


研究了一番
我理解的是国内学者大多用日数据，并且用滞后四个交易日市场收益率做回归
分组算平均 后得到以月为单位的股价信息效率数据

(文章下有附上我参考的文章，是引用李志生老师的文献)


目前我所做的delay1 delay2有些卡住了
用sum的结果是最小值出现了负号
但过往的文献都是正数

在此附上我的指令 希望老师们帮我看看问题

delay1

1. gen mydate=date(tradeweek,"YMD")
   
2. format mydate %tddmy
   
3. gen m = month(mydate)
   
4. gen w = week(mydate)
   
5. rename m month
   
6. rename w week
   
7. order stkcd year month
   
8. 
   
9. *滞后四期
   
10. 
    
11. bys stkcd month : asreg Ri Rm L_Rm L2_Rm  L3_Rm  L4_Rm
    
12. gen R2_δ4=_R2
    
13. drop _*
    
14. 
    
15. bys stkcd month: asreg Ri Rm
    
16. gen R2 = _R2
    
17. drop _*
    
18. gen D1=1-(R2/R2_δ4)
    
19. 
    
20. bysort stkcd year : egen d1=mean(D1)
    
21. 
    
22. sum d1

复制代码

delay2

1. gen mydate=date(tradeweek,"YMD")
   
2. format mydate %tddmy
   
3. gen m = month(mydate)
   
4. gen w = week(mydate)
   
5. rename m month
   
6. rename w week
   
7. order stkcd year month
   
8. 
   
9. bys stkcd month: asreg Ri Rm L_Rm L2_Rm  L3_Rm  L4_Rm
   
10. 
    
11. statsby _b, by(stkcd) saving("b.dta", replace): reg Ri Rm L_Rm L2_Rm  L3_Rm  L4_Rm
    
12. merge m:1 stkcd using "b.dta"
    
13. 
    
14. 
    
15. gen δi = _b_L_Rm+_b_L2_Rm+_b_L3_Rm+_b_L4_Rm
    
16. gen β =abs(_b_Rm)
    
17. gen D2 = δi/(β+δi)
    
18. 
    
19. winsor2 D2,replace cuts(1 99)
    
20. bysort stkcd year : egen d2=mean(D2)
    
21. 
    
22. sum d2

复制代码

1. * Example generated by -dataex-. To install: ssc install dataex
   
2. clear
   
3. input long stkcd str12 tradeweek str21 日个股流通市值 double Ri float(year Rm L_Rm) str6 证券代码 str9 市场类型 str3 行业代码 str12 交易状态 float(N id t L2_Rm L3_Rm L4_Rm) str3 ind
   
4. 2 "2009-01-05" "62914458.27"   .03876 2009  .039851 -.009905 "000002" "深圳A" "K70" "正常交易" 242 1  1        .        .        . "K"
   
5. 2 "2009-01-06" "64792501.8"   .029851 2009  .032594  .039851 "000002" "深圳A" "K70" "正常交易" 242 1  2        .        .        . "K"
   
6. 2 "2009-01-07" "64416893.09" -.005797 2009 -.002036  .032594 "000002" "深圳A" "K70" "正常交易" 242 1  3  .039851        .        . "K"
   
7. 2 "2009-01-08" "64792501.8"   .005831 2009 -.022718 -.002036 "000002" "深圳A" "K70" "正常交易" 242 1  4  .032594  .039851        . "K"
   
8. 2 "2009-01-09" "64698599.62" -.001449 2009  .020454 -.022718 "000002" "深圳A" "K70" "正常交易" 242 1  5 -.002036  .032594  .039851 "K"
   
9. 2 "2009-01-12" "63947382.21" -.011611 2009  .004423  .020454 "000002" "深圳A" "K70" "正常交易" 242 1  6 -.022718 -.002036  .032594 "K"
   
10. 2 "2009-01-13" "61693729.97" -.035242 2009 -.024805  .004423 "000002" "深圳A" "K70" "正常交易" 242 1  7  .020454 -.022718 -.002036 "K"
    
11. 2 "2009-01-14" "64698599.62"  .048706 2009  .042738 -.024805 "000002" "深圳A" "K70" "正常交易" 242 1  8  .004423  .020454 -.022718 "K"
    
12. 2 "2009-01-15" "64322990.92" -.005806 2009   .00339  .042738 "000002" "深圳A" "K70" "正常交易" 242 1  9 -.024805  .004423  .020454 "K"
    
13. 2 "2009-01-16" "64980306.15"  .010219 2009  .014589   .00339 "000002" "深圳A" "K70" "正常交易" 242 1 10  .042738 -.024805  .004423 "K"
    
14. end

复制代码

我的思路是搜集日数据(日个股收益率那些)
再将股票中以月为单位回归 以年为单位求平均
因为最后的话我需要得到每只股票对应每年的股价信息效率

有问题的部份烦请指教

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0 回帖

关键词：replace statsby winsor2 rename bysort