判断混合效应，可用LSDV法
reg y x1 x2 x3 i.state, vce(cluster state)
estimates store LSDV
在结果中看各个state的p值，若大多数为零，则可拒绝“所有个体虚拟变量都为0”的原假设，即存在个体效应，不应使用混合回归。

如果觉得回答还可以，设为最佳答案哦，赚点论坛币

太谢谢您了！！

reg Cscore Overconf Own Salesgrowth Invest CFO Revenue ROA i.Stkcd, vce(cluster Stkcd)
matsize too small
    You have attempted to create a matrix with too many rows or columns or attempted to
    fit a model with too many variables.  You need to increase matsize; it is currently
    400.  Use set matsize; see help matsize.

    If you are using factor variables and included an interaction that has lots of missing
    cells, either increase matsize or set emptycells drop to reduce the required matrix
    size; see help set emptycells.

    If you are using factor variables, you might have accidentally treated a continuous
    variable as a categorical, resulting in lots of categories.  Use the c. operator on
    such variables.
stata 报错了这是什么意思啊

把matsize设定大一些，比如set matsize 1000

. set matsize 1000

. do "C:\Users\Lenovo\AppData\Local\Temp\STD00000000.tmp"

. reg Cscore Overconf Own Salesgrowth Invest CFO Revenue ROA i.Stkcd, vce(cluster Stkcd)
matsize too small
    You have attempted to create a matrix with too many rows or columns or attempted to
    fit a model with too many variables.  You need to increase matsize; it is currently
    1000.  Use set matsize; see help matsize.

    If you are using factor variables and included an interaction that has lots of missing
    cells, either increase matsize or set emptycells drop to reduce the required matrix
    size; see help set emptycells.

    If you are using factor variables, you might have accidentally treated a continuous
    variable as a categorical, resulting in lots of categories.  Use the c. operator on
    such variables.
还是一样的会报错

恩 我又把它射程2000之后就可以了，我想问一下. 下面这个回归出来的结果是表示的什么回归呢？还有  右上角的 F 和P 为什么没有值啊？
reg Cscore Overconf Own Salesgrowth Invest CFO Revenue ROA i.Stkcd, vce(cluster Stkcd)

Linear regression                                      Number of obs =    4110
                                                       F(  6,  1369) =       .
                                                       Prob > F      =       .
                                                       R-squared     =  0.3662
                                                       Root MSE      =  .11034

                               (Std. Err. adjusted for 1370 clusters in Stkcd)
------------------------------------------------------------------------------
             |               Robust
      Cscore |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    Overconf |  -.0140743   .0078471    -1.79   0.073     -.029468    .0013194
         Own |   .2809291   .1400276     2.01   0.045     .0062372     .555621
Salesgrowth |   .0018547   .0063816     0.29   0.771    -.0106642    .0143735
      Invest |   .0489164   .0349428     1.40   0.162    -.0196309    .1174636
         CFO |  -.1302071   .0548429    -2.37   0.018    -.2377923   -.0226218
     Revenue |   .0551777   .0183238     3.01   0.003     .0192319    .0911235
         ROA |  -.2060019   .0881102    -2.34   0.020    -.3788475   -.0331564

这个是LSDV法判断是否存在个体效应和是否需使用混合回归。不需要右上角的P和F值，只需要看表格下面你设定的Stkcd对应的p值，如果大多数p值都接近0，就表示不应使用混合回归

为何我设定为11000 还是会出现这个情况

可以先进行随机效应回归再用命令xttest0检验，同意假设就用混合回归