Dummy variable crisis

[Tommy Duong]

I have a regress model the research about financial crisis. I have a plan to use three dummy. The first ones is before crisis (CR1), the second is crisis (CR2) and the last one is post crisis (CR3). Is it ok? Cuz my teacher said I should use 2 dummy, not nescessary to use three, it can lead to multicollinrearity. What should I do? I dunno how to use that with two dummy. My research I suppose to do like comparation before crisis, crisis and after. I need help!!!

[Carlo Lazzaro]

Tommy:
dummy trap is usually covered in any decent textbook on statistics but can be easily avoided in Stata (see -help fvvarlist-).
As an aside, due to the scant details that you reported about your research project, I would assume that you have cross-sectional data.
That said, as far as your predictor -crisis- is concerned, you may want to consider using a three-level single variable (before crisis; during crisis; after crisis), and choose one out of them as reference level, like in the following toy example:

set obs 100
g income=runiform()*10000
g crisis=1 in 1/30
replace crisis=2 in 31/60
replace crisis=3 if crisis==.
label define crisis 1 before_crisis 2 during_crisis 3 post_crisis
label val crisis crisis
regress income i.crisis, baselevels

As an acid test, you may also want to -summarize. -income- before, during and after crisis via the following code:

bysort crisis: sum income

[Tommy Duong]

LEV= α + β TAN + β2PR + β3FS + β4GR + β5INF + β6FA + β7CRI +β8CRI2 + β9CRI3 + ε
CRI1 b4crisis, CRI2 during crisis, CRI3 after. Does it lead multicorlinear

[Carlo Lazzaro]

Tommy:
your code (for me, at leat) is not that informative.
Please, take a look at FAQ on how to post effectively.

[Devra Golbe]

Tommy,

I second Carlo's advice. In addition, it appears that you seek answers to a class assignment. Although I do not see it explicitly in the forum FAQ, you should be aware that forum members are generally unwilling to do your homework for you.

Devra Golbe
Dept. of Economics
Hunter College, CUNY

[Carlo Lazzaro]

Tommy:
if Devra got it right (and I think she had), please note that seeking for class assignment/homeworks solution is out of the scope of this forum.

[Enoch Ahuakesey]

I want to Run an OLS regression with log wage (ln wage) as the dependent variable controlling for the dummy variable for college graduate and the continuous variable age. I'm confused, any help please?

[Jared Greathouse]

Enoch Ahuakesey

reg ln_wage i.college age

[Enoch Ahuakesey]

Thanks but , @Jared i used what you suggested but dont see any difference between that and reg ln_wage college age

[Enoch Ahuakesey]

I mean in their results

[Carlo Lazzaro]

Enoch:
if you have a two-level categorical variable, prefixing it with -i.- or not does not make any difference (as per -fvvarlist- underlying machinery):

. use "C:\Program Files\Stata17\ado\base\a\auto.dta"
(1978 automobile data)

. reg price foreign mpg

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(2, 71)        =     14.07
       Model |   180261702         2  90130850.8   Prob > F        =    0.0000
    Residual |   454803695        71  6405685.84   R-squared       =    0.2838
-------------+----------------------------------   Adj R-squared   =    0.2637
       Total |   635065396        73  8699525.97   Root MSE        =    2530.9

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
     foreign |   1767.292    700.158     2.52   0.014     371.2169    3163.368
         mpg |  -294.1955   55.69172    -5.28   0.000    -405.2417   -183.1494
       _cons |   11905.42   1158.634    10.28   0.000     9595.164    14215.67
------------------------------------------------------------------------------

. reg price i.foreign mpg

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(2, 71)        =     14.07
       Model |   180261702         2  90130850.8   Prob > F        =    0.0000
    Residual |   454803695        71  6405685.84   R-squared       =    0.2838
-------------+----------------------------------   Adj R-squared   =    0.2637
       Total |   635065396        73  8699525.97   Root MSE        =    2530.9

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
     foreign |
    Foreign  |   1767.292    700.158     2.52   0.014     371.2169    3163.368
         mpg |  -294.1955   55.69172    -5.28   0.000    -405.2417   -183.1494
       _cons |   11905.42   1158.634    10.28   0.000     9595.164    14215.67
------------------------------------------------------------------------------

.

That said, as per FAQ:
1) you should be as informative as possible in your queries (in this case, you did not specify how many levels your categorical variable is composed of);
2) you should post what you typed and what Stata gave you back (via CODE delimiters, please). Thanks.

[Enoch Ahuakesey]

Thanks @ Carlo, I am actually new to stata that's why. Im grateful for the help

[Enoch Ahuakesey]

Union has been classified as 0 and 1 and dots. How do i miss missing union status and use '' miss missing union status" as part of my independent variables?

[Jared Greathouse]

You need to use dataex to show us your data and show us the exact code you've used so far, using the code delimiters. Otherwise, we can't help you Enoch Ahuakesey I don't want you to describe your data to me, I want to see what it really looks like in your dataset, and we can't do that unless you show us your example data. As Carlo said, you have to read the FAQ in order for us to be of any use here.

[Carlo Lazzaro]

Enoch:
we were totally newbies with Stata at the beginning (and as far as the many Stata commands that I do not know, I'm still feel as a beginner).
Unfortunately, this is not a waiver to skip the FAQ (that in the mid/long run will save your time, if followed), as Jared wisely highlighted.