货源通是一个正规货源信息发布平台,发布信息前必须先实名认证,只接受发布货源信息,不可发布违法信息。
免费发布信息
货源通 > 餐饮行业新闻资讯 > 实力工厂货源 > 手表实力厂家 >  京东上的劳力士是高仿表吗盘点劳力士日志复刻手表顶级货 Stata学习:如何复刻一篇实证论文?(六)


  • 【莆田鞋厂家分类】
  • 【奢侈大牌包包厂家分类】
  • 【潮牌奢侈服饰鞋子厂家分类】
  • 【名表厂家分类】

厂家货源分类区域

京东上的劳力士是高仿表吗盘点劳力士日志复刻手表顶级货 Stata学习:如何复刻一篇实证论文?(六)

发布时间:2024-03-31 16:21:20  来源:互联网整理  浏览:   【】【】【

京东上的劳力士是高仿表吗盘点劳力士日志复刻手表顶级货 Stata学习:如何复刻一篇实证论文?(六)

需要找顶级复刻国际名表源头货源 请加下面手表厂家微信咨询,他们是复刻顶级名表源头工厂,主营:顶级名牌名表批发|各种电商合作招代理 一件代发|以及终端零售。

请用微信扫一扫下面二维码加全部厂家咨询,手机端用户:请长按二维码保存到相册后,用微信扫一扫功能读取相册里的二维码可自动加微信!(请全部加他们咨询与对比,并且告诉他们需要找什么品牌具体款式的名表,他们会为你找款和介绍,他们是工厂各种名表应有尽有)

下面给大家讲解“盘点劳力士日志复刻手表顶级货 Stata学习:如何复刻一篇实证论文?(六)”的知识,本站信息仅供大家参考哦!

温馨提示:只有上面二维码是本站推荐的顶级复刻名表源头厂家,分割线以下的其他微信号请不要加,因为文章是网络整理过来的,骗子多请注意!

-------------------分割线以下-----商家微信号请不要联系----------------

 

盘点劳力士日志复刻手表顶级货 Stata学习:如何复刻一篇实证论文?(六) 

盘点劳力士日志复刻手表顶级货

高档精品复刻表高仿手表微信号:6963220

(无需点开,直接加微信咨询)高仿十年老店,目前买的人最多的渠道。

每款宇舶表传世之作新品的发布总是备受宇舶表藏家们的期待,MP-10线性陀传动系统陀飞轮钛金腕表也不例外。全新时计一经发布便获得广泛关注,成为瞩目焦点。圆润且流畅的设计及精致复杂的蓝宝石镜面,并搭载同样精密复杂的机芯,成就出将设计与功能完美融合的典范佳作。全新时计生动诠释了机械装置与美学设计密不可分的理念,而这正是打造传世系列腕表作品的核心。

Norqain Adventure Sport Chrono 44毫米腕表展现了品牌精湛的制表技艺。黑色DLC涂层精钢表壳,直径达到44毫米,表耳宽52毫米,厚度为16毫米,坚固而精致。滚花单向旋转表圈采用相同材质制成,前15分钟有标记,后5分钟有凹槽,倒三角形内有零点,既实用又具有设计感。表壳右侧装饰有一对计时按钮,与旋入式上链表冠相得益彰。左侧则刻有Norqain的标志性铭牌,还可根据个人喜好刻上独特的铭文。表盘采用品牌标志性的蔚蓝色调,与黑色小表盘的对比鲜明,令人眼前一亮。

4. 有关劳力士手表更换表带的小妙招;

5. 教您如何精确调整劳力士手表表带上手长度。

推荐12313132【编号:AHG83736963220】(店铺中不定期更新款式细节)

行业专家提醒大家:“劳力士身为经典品牌,深受广大表迷追捧。市面上仿制n厂的黑水鬼款式虽众,选表务必仔细甄别。”

在众多仿制劳力士黑水鬼手表中,许多消费者对其外观钟爱有加。因厂家在生产阶段特别注重设计,使仿制手表在外观上高度还原原版风貌,展示出绝佳的时尚品位,从而备受广大消费者喜爱。除此以外,表身质量同样取决于生产工艺。仅有严把工艺关,方确保表身稳定性。挑选一款理想的仿真手表对消费者来说可谓是一种挑战,而诸如奢表会这样的热销手表销售平台,无论是商品品种抑或是服务素质均得到良好保证,全力满足大众对于手表的追求。然而购表之时,请务必选择正规途径,以免误买低价劣质产品。只有目标明确,选购质量合格的手表,您才能够得到更加出色的使用体验。

期待各位读者在留言区踊跃发表见解,与笔者进行有意义的交流。更希望你们将本文转发至亲朋好友圈,共同探讨,共享独到的购表心得与实用小技巧。此次大会议题:如何根据个人喜好,挑选心仪的手表,赶快加入讨论吧!

高档精品复刻表高仿手表微信号:6963220(无需点开,直接加微信咨询)高仿十年老店,目前买的人最多的渠道。


Stata学习:如何复刻一篇实证论文?(六)

前情回顾

文献来源

本文考察了中南大学任晓航副教授及其团队在2023年发表在Energy Economics上的“Does carbon price uncertainty affect stock price crash risk? Evidence from China”文章的实证部分的可复现性。

Ren等(2023)探讨碳价不确定性对股价崩盘风险的影响。利用2011-2018年中国上市公司数据的动态面板模型,发现高碳价格不确定性增加了股价崩盘风险。碳价格不确定性的影响在重污染行业和《巴黎协定》后阶段更为突出。碳价格不确定性导致股价暴跌的两个潜在渠道是管理者对坏消息的囤积和投资者的异质性。此外,减少企业内外的信息不对称可以减轻碳价格不确定性对股价崩盘风险的影响。结果表明,碳价格的不确定性作为一个未被充分探索的新因素,由普遍的巨灾风险抑制引起,对股票价格有意想不到的重要影响。

本文发现,Ren等(2023)给出了数据集,但并没有给出代码。因此本文试图用代码还原该论文结果(所有语句均需要放入do文件运行)。需要注意的是,每张表格的复刻方式不唯一,本文仅采用其中一种方法。感兴趣的读者可以设计其他解法予以验证。

数据来源

Ren, X., et al. (2023). Does carbon price uncertainty affect stock price crash risk? Evidence from China

Appendix B. Supplementary data【编号:AHG8373数据】(给出了独立的分析师预测结果、股价崩盘、上市公司数据、碳价波动率、盈余管理结果的dta文件)

合并数据

cd "C:Download1-s2.0-S0140988323001871-mmc1Regression results"

use 上市公司数据, clear

merge m:1 year using 碳价波动率, nogen keep(1 3)

foreach i in 股价崩盘 盈余管理结果 分析师预测结果{

merge 1:1 companyid year using `i', nogen keep(1 3)

}

drop I-L

la var ssrq 上市日期

la var zczj 资产总计

la var zccs 注册城市

la var zcsf 注册省份

la var cp 碳价不确定性

la var lncp 碳价不确定性自然对数

ren companyid Stkcd

compress

save 复刻六, replace

d

得到:

Contains data from 复刻六.dta

Observations: 36,459

Variables: 54 27 Apr 2023 06:12

----------------------------------------------------------------------------------

Variable Storage Display Value

name type format label Variable label

----------------------------------------------------------------------------------

Stkcd double %10.0g companyid

year float %10.0g year

ssrq int %td.. 上市日期

industry str3 %9s industry

gzwzczxj double %10.0g gzwzczxj

gdzcje double %10.0g gdzcje

wxzcje double %10.0g wxzcje

zczj double %10.0g 资产总计

zccs str15 %15s 注册城市

zcsf str9 %9s 注册省份

ipoyear int %10.0g ipoyear

TobinQ double %10.0g TobinQ

Turnover double %10.0g Turnover

Size double %10.0g Size

Lnsize double %10.0g Lnsize

Capex double %10.0g Capex

PPE double %10.0g PPE

Asset_growth double %10.0g Asset_growth

Firm_Age byte %10.0g Firm_Age

Intangible double %10.0g Intangible

HHI double %10.0g HHI

ROE double %10.0g ROE

Growth double %10.0g Growth

EPS double %10.0g EPS

Institutional double %10.0g Institutional

ROA double %10.0g ROA

Duality byte %10.0g Duality

Independent double %10.0g Independent

Top1 double %10.0g Top1

Analysts double %10.0g Analysts

Lev double %10.0g Lev

Cash_Holding double %10.0g Cash_Holding

MTB double %10.0g MTB

tze double %10.0g tze

Investment double %10.0g Investment

cp double %10.0g 碳价不确定性

lncp double %10.0g 碳价不确定性自然对数

Ret double %10.0g Ret

LW byte %10.0g LW

Crash byte %10.0g Crash

sigma double %10.0g sigma

NCSKEW double %10.0g NCSKEW

DUVOL double %10.0g DUVOL

证券代码 str6 %9s 证券代码

行业代码 str3 %9s 行业代码

Industry str3 %9s

DA float %9.0g

AbsDA float %9.0g

accm float %9.0g

实际每股收益 double %10.0g 调整前每股收益

预测每股收益~值 double %10.0g (mean) 预测每股收益

预测每股收益~差 double %10.0g (sd) 预测每股收益

FERROR float %9.0g FERROR

FDISP float %9.0g FDISP

----------------------------------------------------------------------------------

Sorted by: Stkcd year

作者采用的模型是:

图源:Ren等(2023)

作者采用动态面板模型进行分析,并采用系统GMM方法。此外,根据Roodman(2009),作者对回归结果进行了两种检验:

Hansen检验,表明所选工具变量的有效性

AR(2)检验,检验误差项是否具有序列相关性

当两个对应的p值都大于0.1时,这意味着该模型通过了两个检验,并且系统GMM的估计始终有效。

Roodman, D. (2009).How to do Xtabond2: An Introduction to Difference and System GMM in Stata. The Stata Journal: Promoting Communications on Statistics and Stata, 9(1), 86–136.

表1:描述性统计

作者排除了金融行业公司样本中的数据。合并各种数据后,排除了缺少财务数据的公司样本。作者最终样本包括2011年至2018年的13712个样本观测值。还对模型中所有连续变量在1%和99%的水平上缩尾。

图源:Ren等(2023)

但是本文发现,合并后的数据量大于13712:

use 复刻六, clear

ren (lncp Ret sigma Lev Lnsize Analysts Institutional ) (LNCPUN RETPCT SIGMA LEV SIZE ANA INS)

xtset Stkcd year

qui foreach v of var NCSKEW DUVOL RETPCT SIGMA ROA LEV SIZE ANA INS MTB{

g L_`v' = L.`v'

}

keep if !mi(NCSKEW,DUVOL,LNCPUN,L_NCSKEW,L_DUVOL,L_RETPCT,L_SIGMA,L_ROA,L_LEV,L_SIZE,L_ANA,L_INS,

tabstat NCSKEW DUVOL LNCPUN L_NCSKEW L_DUVOL L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB

得到:

Variable | N Mean SD Min p25 p50 p75 Max

-------------+--------------------------------------------------------------------------------

NCSKEW | 19810 -.2771235 .704743 -3.880157 -.6604033 -.2367804 .1420567 4.990688

DUVOL | 19810 -.1098165 .4946463 -2.049016 -.436283 -.1174393 .2071775 2.741572

LNCPUN | 19810 4.407178 .7483846 3.723277 3.89974 4.013631 5.332567 5.787099

L_NCSKEW | 19810 -.2628881 .6980481 -3.880157 -.6459503 -.2239873 .1589879 4.990688

L_DUVOL | 19810 -.0961229 .4927175 -2.049016 -.4211959 -.101272 .2246273 2.741572

L_RETPCT | 19810 -.0013425 .0013938 -.0220737 -.0015775 -.0009526 -.0005791 -.0000333

L_SIGMA | 19810 .0479651 .0205004 .0082437 .0343678 .0440284 .056592 .2149077

L_ROA | 19810 .0461913 .7914227 -3.994411 .014849 .037894 .067159 108.3657

L_LEV | 19810 .4357029 .3243672 -.194698 .24934 .4187985 .595585 16.54535

L_SIZE | 19810 22.65069 1.140996 19.40666 21.8588 22.52518 23.28831 28.72626

L_ANA | 19810 1.588033 1.125046 0 .6931472 1.609438 2.484907 4.204693

L_INS | 19810 6.459863 7.5258 0 1.0352 3.87485 9.3554 75.0517

L_MTB | 19810 2.800466 19.99593 .0471915 .9679868 1.76279 3.079946 2352.941

----------------------------------------------------------------------------------------------

观察到最后的MTB结果中SD较大,推测原文采用了截断,但是数据量还是无法缩小至作者的范围:

use 复刻六, clear

ren (lncp Ret sigma Lev Lnsize Analysts Institutional ) (LNCPUN RETPCT SIGMA LEV SIZE ANA INS)

xtset Stkcd year

winsor2 NCSKEW DUVOL RETPCT SIGMA ROA LEV SIZE ANA INS MTB, replace cuts(1 99) trim

qui foreach v of var NCSKEW DUVOL RETPCT SIGMA ROA LEV SIZE ANA INS MTB{

g L_`v' = L.`v'

}

keep if !mi(NCSKEW,DUVOL,LNCPUN,L_NCSKEW,L_DUVOL,L_RETPCT,L_SIGMA,L_ROA,L_LEV,L_SIZE,L_ANA,L_INS,

tabstat NCSKEW DUVOL LNCPUN L_NCSKEW L_DUVOL L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB

save 复刻六_1, replace

得到:

Variable | N Mean SD Min p25 p50 p75 Max

-------------+--------------------------------------------------------------------------------

NCSKEW | 16869 -.2667043 .607253 -2.169059 -.6383844 -.233904 .1299529 1.565906

DUVOL | 16869 -.1068421 .445346 -1.226733 -.4202564 -.1140197 .1977866 1.167502

LNCPUN | 16869 4.405015 .7499072 3.723277 3.89974 4.013631 5.332567 5.787099

L_NCSKEW | 16869 -.2462107 .6058295 -2.169059 -.6152436 -.2183197 .1519208 1.556334

L_DUVOL | 16869 -.089951 .4462045 -1.226733 -.404428 -.0974893 .2177266 1.167502

L_RETPCT | 16869 -.0012284 .0009676 -.0061994 -.0015256 -.0009442 -.0005843 -.0001433

L_SIGMA | 16869 .0468819 .0172339 .0170754 .0345113 .0438497 .0556688 .1111555

L_ROA | 16869 .0394132 .0506594 -.348528 .014822 .037547 .065658 .207023

L_LEV | 16869 .427431 .2138736 .050548 .251672 .417392 .589179 1.351834

L_SIZE | 16869 22.58802 1.031064 20.12126 21.86095 22.50092 23.22088 25.91643

L_ANA | 16869 1.573307 1.102538 0 .6931472 1.609438 2.484907 3.663562

L_INS | 16869 6.225861 6.646404 0 1.071 3.9396 9.314 34.8292

L_MTB | 16869 2.236044 1.785317 .1830166 .9955618 1.737088 2.930274 11.74164

----------------------------------------------------------------------------------------------

这里还存在其他可能性,不过先按这样做往下做试一试。

表2:相关系数表

图源:Ren等(2023)

use 复刻六_1, clear

gl A = "NCSKEW DUVOL LNCPUN L_NCSKEW L_DUVOL L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB"

asdoc pwcorr $A, star(all) replace

得到:

| NCSKEW DUVOL LNCPUN L_NCSKEW L_DUVOL L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB

-------------+---------------------------------------------------------------------------------------------------------------------

NCSKEW | 1.0000

DUVOL | 0.8250*** 1.0000

LNCPUN | 0.0184** -0.0422*** 1.0000

L_NCSKEW | 0.0488*** 0.0481*** 0.0866*** 1.0000

L_DUVOL | 0.0449*** 0.0648*** 0.0741*** 0.8305*** 1.0000

L_RETPCT | -0.0056 -0.0376*** 0.0749*** 0.0586*** -0.0170** 1.0000

L_SIGMA | 0.0171** 0.0494*** -0.0701*** -0.0382*** 0.0342*** -0.9753*** 1.0000

L_ROA | 0.0304*** -0.0051 0.0632*** 0.0250*** 0.0050 0.0381*** -0.0400*** 1.0000

L_LEV | -0.0555*** -0.0558*** -0.0349*** -0.0662*** -0.0746*** 0.0700*** -0.0737*** -0.4121*** 1.0000

L_SIZE | -0.0794*** -0.1073*** 0.0468*** -0.1211*** -0.1414*** 0.1009*** -0.1312*** 0.0167** 0.4202*** 1.0000

L_ANA | 0.0481*** 0.0093 0.0179** 0.0413*** 0.0089 0.0703*** -0.0712*** 0.3944*** -0.1180*** 0.3330*** 1.0000

L_INS | 0.0703*** 0.0603*** -0.0416*** 0.0896*** 0.0596*** -0.0008 0.0063 0.1928*** 0.0179** 0.3056*** 0.4590*** 1.0000

L_MTB | 0.0794*** 0.1001*** 0.0641*** 0.1007*** 0.1478*** -0.3862*** 0.3929*** 0.2581*** -0.4652*** -0.2516*** 0.0368*** 0.0687*** 1.0000

表3:基准回归

图源:Ren等(2023)

作者发现,AR(2)和Hansen检验结果没有拒绝原假设,说明模型通过了Arellano-Bond序列相关检验和Hansen检验,估计结果可靠。实证结果表明,碳价格不确定性与股价崩盘风险呈正相关。较高的碳价格不确定性增加了企业面临各种不确定性的可能性,导致内部负面消息的积累和外部投资者的分歧增加,增加了崩盘风险。

参考思路:

use 复刻六_1, clear

ta y, g(y)

gl k = "LNCPUN L_NCSKEW L_DUVOL L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB"

gl c1 = "LNCPUN L_NCSKEW L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2-y8"

gl c2 = "LNCPUN L_DUVOL L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2-y8"

gl i = "LNCPUN L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2-y8"

gl p = "tstat bdec(3) tdec(2) addtext(Year_FE, Yes, Firm_FE, Yes)"

xtabond2 NCSKEW $c1, gmm(NCSKEW, lag(2 4)) iv($i) robust

outreg2 using out.doc, ctitle(NCSKEW) keep($k) sortvar($k) $p adds("AR(2)p-value",`e(ar2p)') replace

loc q1 = round(`e(hansenp)',.001)

xtabond2 DUVOL $c2, gmm(DUVOL, lag(2 4)) iv($i) robust

outreg2 using out.doc, ctitle(DUVOL) keep($k) sortvar($k) $p adds("AR(2)p-value",`e(ar2p)')

loc q2 = round(`e(hansenp)',.001)

di `q1' " " `q2'

shellout using `"out.doc"'

得到结果:

. di `q1' " " `q2'

.369 .534

表4:机制检验

图源:Ren等(2023)

基于Chen等(2018),作者使用两步回归来测试坏消息囤积的渠道。具体而言,使用公司信息透明度(ACCM)作为坏消息囤积的代理变量,ACCM是过去三年操纵应计收益的绝对值之和。结果支持坏消息囤积的渠道。

use 复刻六_1, clear

ta y, g(y)

gl k = "LNCPUN L_NCSKEW L_DUVOL L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB"

gl c1 = "LNCPUN L_NCSKEW L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2-y8"

gl c2 = "LNCPUN L_DUVOL L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2-y8"

gl i = "LNCPUN L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2-y8"

gl p = "tstat bdec(3) tdec(2) addtext(Year_FE, Yes, Firm_FE, Yes)"

xtabond2 NCSKEW $c1, gmm(NCSKEW, lag(2 4)) iv($i) robust

outreg2 using out.doc, ctitle(NCSKEW) keep($k) sortvar($k) $p adds("AR(2)p-value",`e(ar2p)') replace

loc q1 = round(`e(hansenp)',.001)

xtabond2 DUVOL $c2, gmm(DUVOL, lag(2 4)) iv($i) robust

outreg2 using out.doc, ctitle(DUVOL) keep($k) sortvar($k) $p adds("AR(2)p-value",`e(ar2p)')

loc q2 = round(`e(hansenp)',.001)

di `q1' " " `q2'

shellout using `"out.doc"'

得到:

其中,第一列的X被omit了:

. xtabond2 ACCM $c0, gmm(ACCM, lag(2 4)) iv($i) robust

Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.

LNCPUN dropped due to collinearity

Warning: Two-step estimated covariance matrix of moments is singular.

Using a generalized inverse to calculate robust weighting matrix for Hansen test.

Difference-in-Sargan/Hansen statistics may be negative.

Dynamic panel-data estimation, one-step system GMM

------------------------------------------------------------------------------

Group variable: Stkcd Number of obs = 8492

Time variable : year Number of groups = 2123

Number of instruments = 32 Obs per group: min = 1

Wald chi2(15) = 49673.45 avg = 4.00

Prob > chi2 = 0.000 max = 7

------------------------------------------------------------------------------

| Robust

ACCM | Coefficient std. err. z P>|z| [95% conf. interval]

-------------+----------------------------------------------------------------

L_ACCM | 1.039487 .0389446 26.69 0.000 .9631565 1.115817

L_RETPCT | -3.757443 4.372431 -0.86 0.390 -12.32725 4.812365

L_SIGMA | -.2146583 .2441114 -0.88 0.379 -.6931078 .2637912

L_ROA | -.039901 .0310086 -1.29 0.198 -.1006767 .0208747

L_LEV | .01587 .0074841 2.12 0.034 .0012016 .0305385

L_SIZE | -.0042123 .0012704 -3.32 0.001 -.0067022 -.0017224

L_ANA | .0015913 .001123 1.42 0.156 -.0006098 .0037924

L_INS | .000111 .0001568 0.71 0.479 -.0001964 .0004183

L_MTB | .0013663 .000903 1.51 0.130 -.0004035 .0031361

y2 | -.0111646 .0038002 -2.94 0.003 -.0186128 -.0037164

y3 | -.0122402 .0039515 -3.10 0.002 -.0199851 -.0044953

y4 | -.0036187 .004419 -0.82 0.413 -.0122798 .0050424

y5 | .0092778 .0045263 2.05 0.040 .0004064 .0181492

y6 | .0052152 .0041023 1.27 0.204 -.0028251 .0132555

y7 | .0047555 .0037097 1.28 0.200 -.0025153 .0120264

_cons | .0778069 .0288386 2.70 0.007 .0212844 .1343294

------------------------------------------------------------------------------

Instruments for first differences equation

Standard

D.(LNCPUN L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2 y3 y4

y5 y6 y7)

GMM-type (missing=0, separate instruments for each period unless collapsed)

L(2/4).ACCM

Instruments for levels equation

Standard

LNCPUN L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2 y3 y4 y5

y6 y7

_cons

GMM-type (missing=0, separate instruments for each period unless collapsed)

DL.ACCM

------------------------------------------------------------------------------

Arellano-Bond test for AR(1) in first differences: z = -13.22 Pr > z = 0.000

Arellano-Bond test for AR(2) in first differences: z = 6.27 Pr > z = 0.000

------------------------------------------------------------------------------

Sargan test of overid. restrictions: chi2(16) = 188.22 Prob > chi2 = 0.000

(Not robust, but not weakened by many instruments.)

Hansen test of overid. restrictions: chi2(16) = 70.84 Prob > chi2 = 0.000

(Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

GMM instruments for levels

Hansen test excluding group: chi2(11) = 38.24 Prob > chi2 = 0.000

Difference (null H = exogenous): chi2(5) = 32.60 Prob > chi2 = 0.000

iv(LNCPUN L_RETPCT L_SIGMA L_ROA L_LEV L_SIZE L_ANA L_INS L_MTB y2 y3 y4 y5 y6 y7)

Hansen test excluding group: chi2(1) = 12.15 Prob > chi2 = 0.000

Difference (null H = exogenous): chi2(15) = 58.70 Prob > chi2 = 0.000

. outreg2 using out.doc, ctitle(ACCM) keep($k) sortvar($k) $p adds("AR(2)p-value",`e(ar2p)') replace

out.doc

dir : seeout

. loc q0 = round(`e(hansenp)',.001)

. xtabond2 NCSKEW $c1, gmm(NCSKEW, lag(2 4)) iv($i) robust

Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.

Warning: Two-step estimated covariance matrix of moments is singular.

Using a generalized inverse to calculate robust weighting matrix for Hansen test.

Difference-in-Sargan/Hansen statistics may be negative.

Dynamic panel-data estimation, one-step system GMM

------------------------------------------------------------------------------

Group variable: Stkcd Number of obs = 8492

Time variable : year Number of groups = 2123

Number of instruments = 32 Obs per group: min = 1

Wald chi2(16) = 2154.66 avg = 4.00

Prob > chi2 = 0.000 max = 7

------------------------------------------------------------------------------

| Robust

NCSKEW | Coefficient std. err. z P>|z| [95% conf. interval]

-------------+----------------------------------------------------------------

ACCM | -.6748858 1.671593 -0.40 0.686 -3.951148 2.601376

L_NCSKEW | .0510511 .0202642 2.52 0.012 .011334 .0907681

L_RETPCT | 83.09846 41.80254 1.99 0.047 1.166985 165.0299

L_SIGMA | 9.087345 2.532713 3.59 0.000 4.123319 14.05137

L_ROA | .3341097 .2373914 1.41 0.159 -.131169 .7993883

L_LEV | .0910197 .1877132 0.48 0.628 -.2768914 .4589307

L_SIZE | -.0527128 .0127525 -4.13 0.000 -.0777073 -.0277183

L_ANA | .0186059 .0097585 1.91 0.057 -.0005204 .0377321

L_INS | .0037956 .0013713 2.77 0.006 .001108 .0064833

L_MTB | .0057416 .011485 0.50 0.617 -.0167686 .0282517

y2 | .0166618 .0336915 0.49 0.621 -.0493723 .082696

责任编辑:
相关文章:
热门阅读排行
© 货源通