Structural break - testing dependent variable only

[Ashvinder SinghChaddha]

Hi all,

I am undertaking a structural break test for my regression mode, but was informed that I should run the test on my dependent variable as that is my variable of interest. My regression model is as follows:

reg logpax loggdp logoil logfx

I am wondering if the correct way to undertake a structural break test for my y variable is through the following code:

regress logpax, vce(robust)

Once done so, I get the following output:

. regress logpax, vce(robust)

Linear regression Number of obs = 20
F(0, 19) = 0.00
Prob > F = .
R-squared = 0.0000
Root MSE = .42064

------------------------------------------------------------------------------
| Robust
logpax | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_cons | 17.87168 .094059 190.01 0.000 17.67481 18.06855
------------------------------------------------------------------------------

Subsequently, I ran this code to find the structural break (given that I have insufficient observations):

forvalues x = 1/30 {
    cap estat sbsingle,  trim(`x')
    if _rc == 198 {
        continue
    }
    local tlevel = `x'
    continue, break
}
estat sbsingle,  trim(`tlevel')

And obtained the following results:

Test for a structural break: Unknown break date

Number of obs = 20

Full sample: 1 - 20
Trimmed sample: 4 - 18
Estimated break date: 2013
Ho: No structural break

Test Statistic p-value
-----------------------------------------------
swald 79.3194 0.0000
-----------------------------------------------
Coefficients included in test: _cons

Wondering if these steps are correct? Thank you.