Closed friendly closed 9 years ago
I could of course add a gp_conf_int = gpar(lty = 3, lwd = ...) argument to control this.
On 2015-10-17 20:30, Michael Friendly wrote:
Copy editors complained about these, but I'm not sure if we can do anything about it. |distplot()| does not allow control of the parameters of the lines for the CIs, using
|grid.segments(RVAL[, 1], RVAL[, 6], RVAL[, 1], RVAL[, 7], default.units = "native", gp = gpar(lty = 3)) |
I don't think they are that bad, but of course the are all drawn 1/2 column widith, so that makes the lines fainter.
— Reply to this email directly or view it on GitHub https://github.com/friendly/VCDR/issues/59.
FH-Prof. Priv.-Doz. Mag. Dr. David Meyer Institut für Wirtschaftsinformatik
Fachhochschule Technikum Wien Höchstädtplatz 5, 1200 Wien T: +43 1 333 40 77-394 F: +43 1 333 40 77-99 394 E: david.meyer@technikum-wien.at I: www.technikum-wien.at
What provokes the problem is the default lty=3, which may be just legacy from my SAS implementation. That's OK for full sized plots, but doesn't scale down well. lty=2 would be a better default, but we might as well add a gpar() argument to control this generally.
If you can add this to vcd::distplot, I'll modify the figure code.
done in rev. 825 (with default lty=2).
David
On 2015-10-18 02:14, Michael Friendly wrote:
What provokes the problem is the default lty=3, which may be just legacy from my SAS implementation. That's OK for full sized plots, but doesn't scale down well. lty=2 would be a better default, but we might as well add a gpar() argument to control this generally.
If you can add this to vcd::distplot, I'll modify the figure code.
— Reply to this email directly or view it on GitHub https://github.com/friendly/VCDR/issues/59#issuecomment-148963007.
FH-Prof. Priv.-Doz. Mag. Dr. David Meyer Institut für Wirtschaftsinformatik
Fachhochschule Technikum Wien Höchstädtplatz 5, 1200 Wien T: +43 1 333 40 77-394 F: +43 1 333 40 77-99 394 E: david.meyer@technikum-wien.at I: www.technikum-wien.at
Great! Re-ran Ch 3, and they look better.
Copy editors complained about these, but I'm not sure if we can do anything about it.
distplot()
does not allow control of the parameters of the lines for the CIs, usingI don't think they are that bad, but of course the are all drawn 1/2 column widith, so that makes the lines fainter.