[Olalere Oluwaseyi]

xtreg ROA CR1 CR2 LR OR MR SIZE GDP c.year#c.year, fe

Fixed-effects (within) regression               Number of obs      =       112
Group variable: Bank                            Number of groups   =        16

R-sq:  within  = 0.7564                         Obs per group: min =         7
       between = 0.2480                                        avg =       7.0
       overall = 0.1068                                        max =         7

                                                F(8,88)            =     34.16
corr(u_i, Xb)  = -0.9813                        Prob > F           =    0.0000

-------------------------------------------------------------------------------
          ROA |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
          CR1 |  -.0417256   .0036029   -11.58   0.000    -.0488857   -.0345655
          CR2 |   .0838742   .0269203     3.12   0.002     .0303757    .1373727
           LR |   .0149466   .0060187     2.48   0.015     .0029857    .0269075
           OR |  -.0840682    .009011    -9.33   0.000    -.1019756   -.0661607
           MR |  -.1001165   .0826243    -1.21   0.229    -.2643149     .064082
         SIZE |    .120439   .4493875     0.27   0.789    -.7726243    1.013502
          GDP |   .0466287   .0123305     3.78   0.000     .0221245     .071133
              |
c.year#c.year |  -1.65e-06   5.18e-07    -3.18   0.002    -2.68e-06   -6.18e-07
              |
        _cons |    6.81094   2.151891     3.17   0.002     2.534508    11.08737
--------------+----------------------------------------------------------------
      sigma_u |  .22550582
      sigma_e |  .02620023
          rho |    .986681   (fraction of variance due to u_i)
-------------------------------------------------------------------------------
F test that all u_i=0:     F(15, 88) =     2.11              Prob > F = 0.0165 

Please Carlo, i just wanted to know how i can include the time variable in the equation form. Do i've to input the coeficient of the time variable in the fixed effect model above place of 't' in the equation? However, I run a Hausman test and it is in favor of Fixed effect.

Thanks,
Olalere

[Carlo Lazzaro]

Olalere:
it sems, from your regression outcome table, that the linear term for -year- was left unreported.
Was it omitted due to collinearity or what?
If you want to investigate a time trend, what you did is enough (provided that you can track back where the linear term for -year- has moved to).
As a closing-out remark the A symbol you should click on to invoke the CODE delimiters is #, no -PHP-.

[William Lisowski]

Thank you for using code delimiters and for showing us your xtreg command.

To include a linear time trend, your command could be

xtreg ROA CR1 CR2 LR OR MR SIZE GDP year, fe

or equivalently

xtreg ROA CR1 CR2 LR OR MR SIZE GDP c.year, fe

To include both a linear and a quadratic time trend, your command would be

xtreg ROA CR1 CR2 LR OR MR SIZE GDP c.year##c.year, fe

The command you used includes only a quadratic time trend; I think Carlo didn't notice that when he commented that the linear term was left unreported.

xtreg ROA CR1 CR2 LR OR MR SIZE GDP c.year#c.year, fe

[Carlo Lazzaro]

Willliam got me!

[Olalere Oluwaseyi]

Thanks a lot Carlo, and William, you have been really helpful. God bless you.