Dear all,
Thanks to Prof. Baum, a new command "qregplot" is available from SSC.
This new command is similar to -grqreg- (by Joao Pedro Azevedo), and will plot coefficients from after the official -qreg-, -bsqreg- and -sqreg- commands.
This command, however, can be thought of as an upgrade on -grqreg- as it can do the following:
- In contrast with -grqreg-, it will work well when using factor notation.
- It provides more flexibility on deciding how the graph works, as one can add options to each component of the figure: the CI (rarea); point estimates (line); the combination of CI and point estimates (two way); and the combination of all coefficients graphs (graph combine)
- Because the hardest part is often the "graph design" (takes time to make it look just right), -qregplot- can store all the quantile regression coefficients in memory (as you would store a regression output), and create graphs directly from those stored results. No need to "redo" the model estimation.
- Lastly, -qregplot- can also be used after other community-contributed commands. On this version, you could use it after:
xtqreg2, qreg2 and ivqreg2 (Machado - Santos Silva)
qrprocess ( Chernozhukov, Fernández-Val, and B. Melly)
mmqreg, rifhdreg, rif(q()) (Rios-Avila)
One last comment.
You will need to update the following commands if you want to use it: xtqreg2, qreg2, ivqreg2, mmqreg
Hope you find it useful
Fernando
Thanks to Prof. Baum, a new command "qregplot" is available from SSC.
This new command is similar to -grqreg- (by Joao Pedro Azevedo), and will plot coefficients from after the official -qreg-, -bsqreg- and -sqreg- commands.
This command, however, can be thought of as an upgrade on -grqreg- as it can do the following:
- In contrast with -grqreg-, it will work well when using factor notation.
- It provides more flexibility on deciding how the graph works, as one can add options to each component of the figure: the CI (rarea); point estimates (line); the combination of CI and point estimates (two way); and the combination of all coefficients graphs (graph combine)
- Because the hardest part is often the "graph design" (takes time to make it look just right), -qregplot- can store all the quantile regression coefficients in memory (as you would store a regression output), and create graphs directly from those stored results. No need to "redo" the model estimation.
- Lastly, -qregplot- can also be used after other community-contributed commands. On this version, you could use it after:
xtqreg2, qreg2 and ivqreg2 (Machado - Santos Silva)
qrprocess ( Chernozhukov, Fernández-Val, and B. Melly)
mmqreg, rifhdreg, rif(q()) (Rios-Avila)
One last comment.
You will need to update the following commands if you want to use it: xtqreg2, qreg2, ivqreg2, mmqreg
Hope you find it useful
Fernando
Comment