Omitted because of collinearity.

[Duy To]

Hello everyone,

I currently have a panel dataset with the ID variable and Year variable. I am missing an Annual Unemployment Rate variable in my data; therefore, I merge my current data with separate data containing the Annual Unemployment. An example of the separate data containing the Annual Unemployment is as follows:

clear
input Year un_rate  
2000 0.0690
2002 0.0565
2003 0.0422
2005 0.0963
end

And I use the m:1 merging technique, with the key variable being Year, to merge my current data with that data.

However, when I run a regression:

reg y x1 x2 i.year un_rate

it returns an error of "omitted because of collinearity." I include "i.year" because I want to include the time dummy variable to capture the year-fixed effect. I see some papers that include the annual unemployment rate and time dummy variable in one regression as well, so I do not understand why I have this problem. Does anyone know why I have this problem since I fail to understand why Year and un_rate have collinearity?

Are there any solutions to this problem? Thank you so much in advance.

[Joseph Coveney]

Have you tried centering one or both of these? Compute the mean of the variable's data (via, for example, summarize var, meanonly) and then subtract it, creating a new centered version of the variable.

Try the unemployment rate first and see whether that eliminates the problem. (You might need to check also against your other two predictors.) If not, then try centering the year, too, but once you do, it will likely need to be included in the model as a continuous predictor.

[Duy To]

Hello again everyone,

So I have tried a couple of things and see how it does not work.

So basically, as long as one of the variables in the list is time-invariant (e.g., gender), and it is in the regression, it keeps returning "omitted because of collinearity." However, I still do not understand the reason behind this issue.

[Duy To]

Hi Joseph, I have tried your suggestion, but the issue persists. :<

[Duy To]

Hi, it's me, again!

So, I have played around with everything, and this is what i got. If the command is like this:

reg y x1 x2 i.year un_rate // it does not work

But, if the command is like this:

reg y x1 x2  un_rate i.year // it works

Are the indicator variables like i.year supposed to be at the end of the command?

[Carlo Lazzaro]

Duy:
but if you actually have a panel dataset, why not going -xtreg,fe-?
In addition, you went -regress- but did you included -i.panelvar- as a predictor?
In addition, I cannot replicate your problem:

. use "https://www.stata-press.com/data/r17/nlswork.dta"
(National Longitudinal Survey of Young Women, 14-24 years old in 1968)

. reg ln_wage age i.year i.idcode if idcode<=3

      Source |       SS           df       MS      Number of obs   =        39
-------------+----------------------------------   F(17, 21)       =      2.68
       Model |  3.54194923        17  .208349955   Prob > F        =    0.0171
    Residual |  1.63378973        21  .077799511   R-squared       =    0.6843
-------------+----------------------------------   Adj R-squared   =    0.4288
       Total |  5.17573896        38  .136203657   Root MSE        =    .27893

------------------------------------------------------------------------------
     ln_wage | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         age |   .3010572   .3561559     0.85   0.407    -.4396095    1.041724
             |
        year |
         69  |  -.0920902   .5314565    -0.17   0.864    -1.197315    1.013134
         70  |  -.8648493    .779214    -1.11   0.280    -2.485314    .7556149
         71  |  -1.248506    1.09967    -1.14   0.269    -3.535396    1.038383
         72  |   -1.39387   1.443494    -0.97   0.345    -4.395779     1.60804
         73  |  -1.520276    1.79214    -0.85   0.406    -5.247236    2.206684
         75  |  -2.049717   2.495803    -0.82   0.421    -7.240024     3.14059
         77  |  -2.657565   3.203292    -0.83   0.416    -9.319175    4.004045
         78  |  -2.751196   3.557758    -0.77   0.448    -10.14996    4.647567
         80  |  -3.324016   4.267534    -0.78   0.445    -12.19884    5.550808
         82  |  -4.027975   4.983977    -0.81   0.428    -14.39272    6.336774
         83  |  -4.207353   5.333467    -0.79   0.439     -15.2989    6.884199
         85  |  -4.730657   6.044586    -0.78   0.443    -17.30106    7.839747
         87  |  -5.407995   6.755956    -0.80   0.432    -19.45777    8.641785
         88  |  -5.901929   7.348904    -0.80   0.431    -21.18481    9.380954
             |
      idcode |
          2  |  -.3898423     .11632    -3.35   0.003     -.631743   -.1479415
          3  |  -2.247118   2.111457    -1.06   0.299    -6.638133    2.143897
             |
       _cons |  -2.882579   5.734884    -0.50   0.620    -14.80892    9.043766
------------------------------------------------------------------------------

. reg ln_wage age i.idcode i.year if idcode<=3

      Source |       SS           df       MS      Number of obs   =        39
-------------+----------------------------------   F(17, 21)       =      2.68
       Model |  3.54194923        17  .208349955   Prob > F        =    0.0171
    Residual |  1.63378973        21  .077799511   R-squared       =    0.6843
-------------+----------------------------------   Adj R-squared   =    0.4288
       Total |  5.17573896        38  .136203657   Root MSE        =    .27893

------------------------------------------------------------------------------
     ln_wage | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         age |   .3010572   .3561559     0.85   0.407    -.4396095    1.041724
             |
      idcode |
          2  |  -.3898423     .11632    -3.35   0.003     -.631743   -.1479415
          3  |  -2.247118   2.111457    -1.06   0.299    -6.638133    2.143897
             |
        year |
         69  |  -.0920902   .5314565    -0.17   0.864    -1.197315    1.013134
         70  |  -.8648493    .779214    -1.11   0.280    -2.485314    .7556149
         71  |  -1.248506    1.09967    -1.14   0.269    -3.535396    1.038383
         72  |   -1.39387   1.443494    -0.97   0.345    -4.395779     1.60804
         73  |  -1.520276    1.79214    -0.85   0.406    -5.247236    2.206684
         75  |  -2.049717   2.495803    -0.82   0.421    -7.240024     3.14059
         77  |  -2.657565   3.203292    -0.83   0.416    -9.319175    4.004045
         78  |  -2.751196   3.557758    -0.77   0.448    -10.14996    4.647567
         80  |  -3.324016   4.267534    -0.78   0.445    -12.19884    5.550808
         82  |  -4.027975   4.983977    -0.81   0.428    -14.39272    6.336774
         83  |  -4.207353   5.333467    -0.79   0.439     -15.2989    6.884199
         85  |  -4.730657   6.044586    -0.78   0.443    -17.30106    7.839747
         87  |  -5.407995   6.755956    -0.80   0.432    -19.45777    8.641785
         88  |  -5.901929   7.348904    -0.80   0.431    -21.18481    9.380954
             |
       _cons |  -2.882579   5.734884    -0.50   0.620    -14.80892    9.043766
------------------------------------------------------------------------------

.

[Duy To]

Hi Carlo,

Thank you for your reply. It is just my habit that I go model by model from the most basic one (e.g., - reg-) to the more advanced one (e.g., -xtreg-) to see if there are any abnormal things presenting (e.g., unreasonable signs of coefficients etc.). But everything is good now. Thank you for your reply again.

[Carlo Lazzaro]

Duy:
I see your point.
But your -regress- code did not show a panel data regression with -fe- specification, as you -i.panelvar- was not included among the predictors.
More substantively, happy with reading that everything is fine now.
It would also be interesting to see what the problem was and how you fixed it. Thanks.