用EGARCH(1，1)模型估计股票收益波动性时无法收敛~

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

我想利用F-F3模型和EGARCH（1）方法估计若干公司的股票收益波动性，代码如下：

by stkcd, sort : arch r rmrf smb hml, ar(1) earch(1) egarch(1)

在运行前已经tsset，并通过tsfill等命令运用线性插值原理补充了gaps，但是，

有的公司能够收敛，有的公司却长时间不能收敛，其中一个公司的运行结果如下：应该怎么解决呢？

(setting optimization to BHHH)
Iteration 0:   log likelihood = -110.89652  
Iteration 1:   log likelihood =  79.535917  
Iteration 2:   log likelihood =  94.557979  
Iteration 3:   log likelihood =  103.62074  
Iteration 4:   log likelihood =  105.80565  
(switching optimization to BFGS)
Iteration 5:   log likelihood =  109.48566  
Iteration 6:   log likelihood =  111.15653  
Iteration 7:   log likelihood =  112.66912  
Iteration 8:   log likelihood =  112.69316  
Iteration 9:   log likelihood =  113.72097  
Iteration 10:  log likelihood =  114.63639  
Iteration 11:  log likelihood =   115.3768  
Iteration 12:  log likelihood =  115.80462  
Iteration 13:  log likelihood =  115.99594  
Iteration 14:  log likelihood =   116.1957  
(switching optimization to BHHH)
Iteration 15:  log likelihood =  116.39695  
Iteration 16:  log likelihood =  116.52964  
Iteration 17:  log likelihood =  116.67106  
Iteration 18:  log likelihood =  116.78666  
Iteration 19:  log likelihood =  116.86471  
(switching optimization to BFGS)
Iteration 20:  log likelihood =  116.87991  
Iteration 21:  log likelihood =  117.37265  
Iteration 22:  log likelihood =  117.57938  
Iteration 23:  log likelihood =  118.21619  
Iteration 24:  log likelihood =  118.22354  (backed up)
Iteration 25:  log likelihood =  118.34794  
Iteration 26:  log likelihood =  118.37384  (backed up)
Iteration 27:  log likelihood =  118.51367  
Iteration 28:  log likelihood =  118.66599  
Iteration 29:  log likelihood =  119.63453  
(switching optimization to BHHH)
Iteration 30:  log likelihood =  119.71734  (backed up)
Iteration 31:  log likelihood =  119.86386  
Iteration 32:  log likelihood =  119.93108  
Iteration 33:  log likelihood =  121.39321  
Iteration 34:  log likelihood =  121.40142  (backed up)
(switching optimization to BFGS)
Iteration 35:  log likelihood =  121.40856  (backed up)
Iteration 36:  log likelihood =  121.52153  (backed up)
Iteration 37:  log likelihood =  121.56463  (backed up)
Iteration 38:  log likelihood =  121.57978  (backed up)
Iteration 39:  log likelihood =  121.60684  
Iteration 40:  log likelihood =  121.77795  (backed up)
Iteration 41:  log likelihood =  121.78593  
BFGS stepping has contracted, resetting BFGS Hessian (0)
Iteration 42:  log likelihood =  121.79939  
Iteration 43:  log likelihood =  121.80473  (backed up)
Iteration 44:  log likelihood =  121.87265  (backed up)
(switching optimization to BHHH)
Iteration 45:  log likelihood =  121.90267  (not concave)
Iteration 46:  log likelihood =  121.90327  (not concave)
Iteration 47:  log likelihood =   121.9033  (not concave)
Iteration 48:  log likelihood =   121.9033  (not concave)
Iteration 49:  log likelihood =   121.9033  (not concave)
(switching optimization to BFGS)
Iteration 50:  log likelihood =   121.9033  
Iteration 51:  log likelihood =  121.90687  (backed up)
Iteration 52:  log likelihood =  121.91176  
BFGS stepping has contracted, resetting BFGS Hessian (1)
Iteration 53:  log likelihood =  121.96561  
Iteration 54:  log likelihood =  121.97547  (backed up)
Iteration 55:  log likelihood =  121.97892  (backed up)
Iteration 56:  log likelihood =  121.99688  (backed up)
Iteration 57:  log likelihood =  122.01067  (backed up)
Iteration 58:  log likelihood =   122.0796  (backed up)
Iteration 59:  log likelihood =  122.08083  (backed up)
(switching optimization to BHHH)
Iteration 60:  log likelihood =  122.08343  (not concave)
Iteration 61:  log likelihood =   122.0989  (not concave)
Iteration 62:  log likelihood =  122.09904  (not concave)
Iteration 63:  log likelihood =  122.09905  (not concave)
Iteration 64:  log likelihood =  122.09905  (not concave)
(switching optimization to BFGS)
Iteration 65:  log likelihood =  122.09907  
Iteration 66:  log likelihood =  122.10249  (backed up)
Iteration 67:  log likelihood =  122.10632  
BFGS stepping has contracted, resetting BFGS Hessian (2)
Iteration 68:  log likelihood =  122.13144  
Iteration 69:  log likelihood =  122.13217  (backed up)
Iteration 70:  log likelihood =  122.13465  (backed up)
Iteration 71:  log likelihood =  122.14583  (backed up)
Iteration 72:  log likelihood =  122.17706  (backed up)
Iteration 73:  log likelihood =  122.18716  (backed up)
Iteration 74:  log likelihood =   122.1893  (backed up)
(switching optimization to BHHH)
Iteration 75:  log likelihood =   122.1922  (not concave)
Iteration 76:  log likelihood =  122.19523  (not concave)
Iteration 77:  log likelihood =  122.19551  (not concave)
Iteration 78:  log likelihood =  122.19551  (not concave)
Iteration 79:  log likelihood =  122.19552  (not concave)
(switching optimization to BFGS)
Iteration 80:  log likelihood =  122.19552  
Iteration 81:  log likelihood =  122.19599  (backed up)
Iteration 82:  log likelihood =  122.19836  
Iteration 83:  log likelihood =  122.24419  
Iteration 84:  log likelihood =  122.26252  (backed up)
BFGS stepping has contracted, resetting BFGS Hessian (3)
Iteration 85:  log likelihood =   122.2907  
Iteration 86:  log likelihood =  122.29126  (backed up)
Iteration 87:  log likelihood =  122.29439  (backed up)
Iteration 88:  log likelihood =  122.30381  (backed up)
Iteration 89:  log likelihood =  122.31289  (backed up)
(switching optimization to BHHH)
Iteration 90:  log likelihood =  122.32883  (not concave)
Iteration 91:  log likelihood =  122.32915  (not concave)
Iteration 92:  log likelihood =  122.32932  (not concave)
Iteration 93:  log likelihood =  122.32933  (not concave)
Iteration 94:  log likelihood =  122.32933  (not concave)
(switching optimization to BFGS)
Iteration 95:  log likelihood =  122.32934  
Iteration 96:  log likelihood =  122.33342  (backed up)
Iteration 97:  log likelihood =   122.3372  
Iteration 98:  log likelihood =  122.33861  
BFGS stepping has contracted, resetting BFGS Hessian (4)
Iteration 99:  log likelihood =  122.35158  
Iteration 100: log likelihood =  122.35185  (backed up)
Iteration 101: log likelihood =  122.35438  (backed up)
Iteration 102: log likelihood =   122.3594  (backed up)
Iteration 103: log likelihood =  122.37071  (backed up)
Iteration 104: log likelihood =  122.37282  (backed up)
(switching optimization to BHHH)
Iteration 105: log likelihood =  122.37963  (not concave)
Iteration 106: log likelihood =  122.38798  (not concave)
Iteration 107: log likelihood =  122.38838  (not concave)
Iteration 108: log likelihood =  122.38838  (not concave)
Iteration 109: log likelihood =  122.38838  (not concave)
(switching optimization to BFGS)
Iteration 110: log likelihood =  122.38838  
Iteration 111: log likelihood =  122.39197  (backed up)
Iteration 112: log likelihood =  122.39668  
Iteration 113: log likelihood =  122.39697  
Iteration 114: log likelihood =  122.41044  
Iteration 115: log likelihood =  122.41636  
BFGS stepping has contracted, resetting BFGS Hessian (5)
Iteration 116: log likelihood =  122.42216  
Iteration 117: log likelihood =  122.42237  (backed up)
Iteration 118: log likelihood =  122.42915  (backed up)
BFGS stepping has contracted, resetting BFGS Hessian (6)
Iteration 119: log likelihood =  122.43428  (backed up)
(switching optimization to BHHH)
Iteration 120: log likelihood =  122.43428  (not concave)
Iteration 121: log likelihood =  122.43432  (not concave)
Iteration 122: log likelihood =  122.43435  (not concave)
Iteration 123: log likelihood =  122.43435  (not concave)
Iteration 124: log likelihood =  122.43435  (not concave)
(switching optimization to BFGS)
Iteration 125: log likelihood =  122.43435  
Iteration 126: log likelihood =  122.43758  (backed up)

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：EGARCH GARCH 模型估计 ARCH 股票收益 模型 收益

检查原始数据 看有没有出错的地方 看看有没有无效的0