Interpreting quantile regressions results

[P.Suiker]

Hello everone!

I'm looking at the differences in taxable income between wage workers and self-employers.
Therefore, I analyzed the panel data with Pooled OLS and a quantile regression.
In the picture below, there is a part of the results for self-employers were OLS represents the coefficient for the OLS regression and the 0.25 0.50 and 0.75 are the coefficients for the quantile regression
The numbers between the brackets are the clustered standard errors and the numbers with a * are signficiant (p-value < 0.05)

Variable	OLS	.25	.50	.75
Constant	-198917 (102227)	130340 (33090)	130340 (71321)	-497139 (761372
Age	27260* (10811)	1848 (3390)	-7750 (6912)	61706 (60644)
Agesq	-248 (215)	5.9 (74)	65.6 (5541)	-688 (693)
Work_experience	-19725* (5956)	-1735 (1468)	5669 (56)	-49161 (39193)

My question is: What can I say about the quantile regression results of for example age?

Thanks in forward for your message.

Patrick Suiker

[Maarten Buis]

The interpretation is exactly the same as in linear regression (OLS is the name of the algorithm used in finding the estimates, linear regression is the name of the model that you estimate with that algorithm), except that you are explaining the conditional 25th, 50th, and 75th percentiles rather than the conditional mean.