Time Series Regression, how to regress only on a certain period

[Lea Birm]

Hello,

I am currently working with quarterly time series data from 1990 to 2019.

However the Chow Test indicates that I have a structural break in 2013q1. Therefore I want to analyze how my results and coefficients vary if I use the two sub periods 1990-2013q1 2013q2-2019q4.
I use the Newey west HAC Standard error:

Newey c y, lag (4) // c = consumption, y = income
Now I have tried the command if tin (2013q2, 2019q4) and I generated a dummy variable but both variations do not work and I always get the output:

time variable is not regularly spaced

Why does this matter when applying Newey west standard errors but not by the Usual OLS Regression?
And is there a way that I can use an alternative command?

I really apprecite your help,
Thank you!

[Phil Bromiley]

You will increase your chances of useful answer by following the FAQ on asking questions – provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

Ordinary OLS doesn't care about the differences in the time variable since time does not figure into the calculations but any estimate that needs to do serial correlation needs to worry about the time variable.

There is no way to diagnose a problem that appears to be your time variable is not regularly spaced without the data. The first thing to do is to see if the time variable is regularly spaced.