package plotrix: how to account for two two z categories

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

package plotrix: how to account for two two z categories

Luigi
Hello,
I have an experiment measuring optical density (OD) when comparing
three parameters:
a) concentration of the target
b) concentration of the probe
c) concentration of the reporter antibody.
Using plotrix I can nicely draw the first two into clusters, but I
can't get separation for the third parameter. is there a way in
plotrix to custer data according to two, let's say, z parameters (I
call the second high-level parameter as w)? For instance, two
clusters, each separated into two subclusters. Or is this more a job
for lattice?
Thank you.

```
x = rep(c(rep("1000 pmol", 4), rep("0 pmol", 4)),2)
z =  rep(c(rep("Treated", 2), rep("Untreated", 2), rep("Treated", 2),
rep("Untreated", 2)))
w = c(rep("1:1000", 8), rep("1:2000", 8))
y = c(1.18, 1.22, 1.52, 2.37, 0.97, 1.08, 1.17, 1.58, 1.16, 1.16, 2.27, 2.24,
      1.26, 1.28, 1.39, 1.77)
Q = data.frame(x, z, w, y, stringsAsFactors = FALSE)
names(Q) = c("[Target]", "Enzyme", "[Antibody]", "OD")
names(Q) = c("x", "z", "w", "y") # for ease, I'll use single letters
library(plotrix)
brkdn.plot(y~x+z, data=Q,
           pch = c(1, 16), cex = 1.5, type="p",
           main="Single Measurement",
           xlab=expression(bold("S1 nuclease")),
           ylab=expression(bold("Optical density")))
brkdn.plot(y~x+z+w, data=Q,
           pch = c(1, 16), cex = 1.5, type="p",
           main="Double Measurement",
           xlab=expression(bold("S1 nuclease")),
           ylab=expression(bold("Optical density")))
```

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: package plotrix: how to account for two two z categories

Jim Lemon-4
Hi Luigi,
To display a nested breakdown like this I would suggest barNest. This
is one way to display the nesting. Note that if you change the order
of the factors in the formula you will get a different plot, so think
about how you want the summaries nested. Error bars can only be
displayed on the final breakdown.

# because you have fairly long labels, use a wide plot
x11(width=10)
barcol<-list("lightgray",c("#ff00ff","#ff99ff"),
 c("#00ff00","#66ff66"),c("#aaaa00","#888800"))
barNest(y~x+z+w,Q,main="Double Measurement",
 ylab="Response",col=barcol,errbars=TRUE)

Jim

On Wed, Sep 23, 2020 at 10:48 PM Luigi Marongiu
<[hidden email]> wrote:

>
> Hello,
> I have an experiment measuring optical density (OD) when comparing
> three parameters:
> a) concentration of the target
> b) concentration of the probe
> c) concentration of the reporter antibody.
> Using plotrix I can nicely draw the first two into clusters, but I
> can't get separation for the third parameter. is there a way in
> plotrix to custer data according to two, let's say, z parameters (I
> call the second high-level parameter as w)? For instance, two
> clusters, each separated into two subclusters. Or is this more a job
> for lattice?
> Thank you.
>
> ```
> x = rep(c(rep("1000 pmol", 4), rep("0 pmol", 4)),2)
> z =  rep(c(rep("Treated", 2), rep("Untreated", 2), rep("Treated", 2),
> rep("Untreated", 2)))
> w = c(rep("1:1000", 8), rep("1:2000", 8))
> y = c(1.18, 1.22, 1.52, 2.37, 0.97, 1.08, 1.17, 1.58, 1.16, 1.16, 2.27, 2.24,
>       1.26, 1.28, 1.39, 1.77)
> Q = data.frame(x, z, w, y, stringsAsFactors = FALSE)
> names(Q) = c("[Target]", "Enzyme", "[Antibody]", "OD")
> names(Q) = c("x", "z", "w", "y") # for ease, I'll use single letters
> library(plotrix)
> brkdn.plot(y~x+z, data=Q,
>            pch = c(1, 16), cex = 1.5, type="p",
>            main="Single Measurement",
>            xlab=expression(bold("S1 nuclease")),
>            ylab=expression(bold("Optical density")))
> brkdn.plot(y~x+z+w, data=Q,
>            pch = c(1, 16), cex = 1.5, type="p",
>            main="Double Measurement",
>            xlab=expression(bold("S1 nuclease")),
>            ylab=expression(bold("Optical density")))
> ```
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: package plotrix: how to account for two two z categories

Jim Lemon-4
Hi Luigi,
I thought a lot about that when I was writing the function. The only
way I could think of to show the nesting was dots with horizontal
lines and it looked messy and was quite hard to visualize the nesting.
If you do have any great ideas I always welcome contributions to
plotrix.

Jim

On Thu, Sep 24, 2020 at 6:25 PM Luigi Marongiu <[hidden email]> wrote:

>
> Thank you Jim, that is really nice!
> But is there a way to use dots instead of boxes? and how do I control
> the colours?
> Best regards
> Luigi
>
> On Thu, Sep 24, 2020 at 9:29 AM Jim Lemon <[hidden email]> wrote:
> >
> > Hi Luigi,
> > To display a nested breakdown like this I would suggest barNest. This
> > is one way to display the nesting. Note that if you change the order
> > of the factors in the formula you will get a different plot, so think
> > about how you want the summaries nested. Error bars can only be
> > displayed on the final breakdown.
> >
> > # because you have fairly long labels, use a wide plot
> > x11(width=10)
> > barcol<-list("lightgray",c("#ff00ff","#ff99ff"),
> >  c("#00ff00","#66ff66"),c("#aaaa00","#888800"))
> > barNest(y~x+z+w,Q,main="Double Measurement",
> >  ylab="Response",col=barcol,errbars=TRUE)
> >
> > Jim
> >
> > On Wed, Sep 23, 2020 at 10:48 PM Luigi Marongiu
> > <[hidden email]> wrote:
> > >
> > > Hello,
> > > I have an experiment measuring optical density (OD) when comparing
> > > three parameters:
> > > a) concentration of the target
> > > b) concentration of the probe
> > > c) concentration of the reporter antibody.
> > > Using plotrix I can nicely draw the first two into clusters, but I
> > > can't get separation for the third parameter. is there a way in
> > > plotrix to custer data according to two, let's say, z parameters (I
> > > call the second high-level parameter as w)? For instance, two
> > > clusters, each separated into two subclusters. Or is this more a job
> > > for lattice?
> > > Thank you.
> > >
> > > ```
> > > x = rep(c(rep("1000 pmol", 4), rep("0 pmol", 4)),2)
> > > z =  rep(c(rep("Treated", 2), rep("Untreated", 2), rep("Treated", 2),
> > > rep("Untreated", 2)))
> > > w = c(rep("1:1000", 8), rep("1:2000", 8))
> > > y = c(1.18, 1.22, 1.52, 2.37, 0.97, 1.08, 1.17, 1.58, 1.16, 1.16, 2.27, 2.24,
> > >       1.26, 1.28, 1.39, 1.77)
> > > Q = data.frame(x, z, w, y, stringsAsFactors = FALSE)
> > > names(Q) = c("[Target]", "Enzyme", "[Antibody]", "OD")
> > > names(Q) = c("x", "z", "w", "y") # for ease, I'll use single letters
> > > library(plotrix)
> > > brkdn.plot(y~x+z, data=Q,
> > >            pch = c(1, 16), cex = 1.5, type="p",
> > >            main="Single Measurement",
> > >            xlab=expression(bold("S1 nuclease")),
> > >            ylab=expression(bold("Optical density")))
> > > brkdn.plot(y~x+z+w, data=Q,
> > >            pch = c(1, 16), cex = 1.5, type="p",
> > >            main="Double Measurement",
> > >            xlab=expression(bold("S1 nuclease")),
> > >            ylab=expression(bold("Optical density")))
> > > ```
> > >
> > > ______________________________________________
> > > [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> > > https://stat.ethz.ch/mailman/listinfo/r-help
> > > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > > and provide commented, minimal, self-contained, reproducible code.
>
>
>
> --
> Best regards,
> Luigi

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: package plotrix: how to account for two two z categories

Jim Lemon-4
Oh, sorry, forgot about the colors. A list beginning with the color
for the overall summary, then colors for the first factor and so on.
See the help page for examples.

Jim

On Thu, Sep 24, 2020 at 6:32 PM Jim Lemon <[hidden email]> wrote:

>
> Hi Luigi,
> I thought a lot about that when I was writing the function. The only
> way I could think of to show the nesting was dots with horizontal
> lines and it looked messy and was quite hard to visualize the nesting.
> If you do have any great ideas I always welcome contributions to
> plotrix.
>
> Jim
>
> On Thu, Sep 24, 2020 at 6:25 PM Luigi Marongiu <[hidden email]> wrote:
> >
> > Thank you Jim, that is really nice!
> > But is there a way to use dots instead of boxes? and how do I control
> > the colours?
> > Best regards
> > Luigi
> >
> > On Thu, Sep 24, 2020 at 9:29 AM Jim Lemon <[hidden email]> wrote:
> > >
> > > Hi Luigi,
> > > To display a nested breakdown like this I would suggest barNest. This
> > > is one way to display the nesting. Note that if you change the order
> > > of the factors in the formula you will get a different plot, so think
> > > about how you want the summaries nested. Error bars can only be
> > > displayed on the final breakdown.
> > >
> > > # because you have fairly long labels, use a wide plot
> > > x11(width=10)
> > > barcol<-list("lightgray",c("#ff00ff","#ff99ff"),
> > >  c("#00ff00","#66ff66"),c("#aaaa00","#888800"))
> > > barNest(y~x+z+w,Q,main="Double Measurement",
> > >  ylab="Response",col=barcol,errbars=TRUE)
> > >
> > > Jim
> > >
> > > On Wed, Sep 23, 2020 at 10:48 PM Luigi Marongiu
> > > <[hidden email]> wrote:
> > > >
> > > > Hello,
> > > > I have an experiment measuring optical density (OD) when comparing
> > > > three parameters:
> > > > a) concentration of the target
> > > > b) concentration of the probe
> > > > c) concentration of the reporter antibody.
> > > > Using plotrix I can nicely draw the first two into clusters, but I
> > > > can't get separation for the third parameter. is there a way in
> > > > plotrix to custer data according to two, let's say, z parameters (I
> > > > call the second high-level parameter as w)? For instance, two
> > > > clusters, each separated into two subclusters. Or is this more a job
> > > > for lattice?
> > > > Thank you.
> > > >
> > > > ```
> > > > x = rep(c(rep("1000 pmol", 4), rep("0 pmol", 4)),2)
> > > > z =  rep(c(rep("Treated", 2), rep("Untreated", 2), rep("Treated", 2),
> > > > rep("Untreated", 2)))
> > > > w = c(rep("1:1000", 8), rep("1:2000", 8))
> > > > y = c(1.18, 1.22, 1.52, 2.37, 0.97, 1.08, 1.17, 1.58, 1.16, 1.16, 2.27, 2.24,
> > > >       1.26, 1.28, 1.39, 1.77)
> > > > Q = data.frame(x, z, w, y, stringsAsFactors = FALSE)
> > > > names(Q) = c("[Target]", "Enzyme", "[Antibody]", "OD")
> > > > names(Q) = c("x", "z", "w", "y") # for ease, I'll use single letters
> > > > library(plotrix)
> > > > brkdn.plot(y~x+z, data=Q,
> > > >            pch = c(1, 16), cex = 1.5, type="p",
> > > >            main="Single Measurement",
> > > >            xlab=expression(bold("S1 nuclease")),
> > > >            ylab=expression(bold("Optical density")))
> > > > brkdn.plot(y~x+z+w, data=Q,
> > > >            pch = c(1, 16), cex = 1.5, type="p",
> > > >            main="Double Measurement",
> > > >            xlab=expression(bold("S1 nuclease")),
> > > >            ylab=expression(bold("Optical density")))
> > > > ```
> > > >
> > > > ______________________________________________
> > > > [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> > > > https://stat.ethz.ch/mailman/listinfo/r-help
> > > > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > > > and provide commented, minimal, self-contained, reproducible code.
> >
> >
> >
> > --
> > Best regards,
> > Luigi

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: package plotrix: how to account for two two z categories

Jim Lemon-4
Hi Luigi,
Here is a quick example of how points would look in the same
configuration. Perhaps with end caps to show the nesting it may be
more like what you want.

Jim

On Thu, Sep 24, 2020 at 6:34 PM Jim Lemon <[hidden email]> wrote:

>
> Oh, sorry, forgot about the colors. A list beginning with the color
> for the overall summary, then colors for the first factor and so on.
> See the help page for examples.
>
> Jim
>
> On Thu, Sep 24, 2020 at 6:32 PM Jim Lemon <[hidden email]> wrote:
> >
> > Hi Luigi,
> > I thought a lot about that when I was writing the function. The only
> > way I could think of to show the nesting was dots with horizontal
> > lines and it looked messy and was quite hard to visualize the nesting.
> > If you do have any great ideas I always welcome contributions to
> > plotrix.
> >
> > Jim
> >
> > On Thu, Sep 24, 2020 at 6:25 PM Luigi Marongiu <[hidden email]> wrote:
> > >
> > > Thank you Jim, that is really nice!
> > > But is there a way to use dots instead of boxes? and how do I control
> > > the colours?
> > > Best regards
> > > Luigi
> > >
> > > On Thu, Sep 24, 2020 at 9:29 AM Jim Lemon <[hidden email]> wrote:
> > > >
> > > > Hi Luigi,
> > > > To display a nested breakdown like this I would suggest barNest. This
> > > > is one way to display the nesting. Note that if you change the order
> > > > of the factors in the formula you will get a different plot, so think
> > > > about how you want the summaries nested. Error bars can only be
> > > > displayed on the final breakdown.
> > > >
> > > > # because you have fairly long labels, use a wide plot
> > > > x11(width=10)
> > > > barcol<-list("lightgray",c("#ff00ff","#ff99ff"),
> > > >  c("#00ff00","#66ff66"),c("#aaaa00","#888800"))
> > > > barNest(y~x+z+w,Q,main="Double Measurement",
> > > >  ylab="Response",col=barcol,errbars=TRUE)
> > > >
> > > > Jim
> > > >
> > > > On Wed, Sep 23, 2020 at 10:48 PM Luigi Marongiu
> > > > <[hidden email]> wrote:
> > > > >
> > > > > Hello,
> > > > > I have an experiment measuring optical density (OD) when comparing
> > > > > three parameters:
> > > > > a) concentration of the target
> > > > > b) concentration of the probe
> > > > > c) concentration of the reporter antibody.
> > > > > Using plotrix I can nicely draw the first two into clusters, but I
> > > > > can't get separation for the third parameter. is there a way in
> > > > > plotrix to custer data according to two, let's say, z parameters (I
> > > > > call the second high-level parameter as w)? For instance, two
> > > > > clusters, each separated into two subclusters. Or is this more a job
> > > > > for lattice?
> > > > > Thank you.
> > > > >
> > > > > ```
> > > > > x = rep(c(rep("1000 pmol", 4), rep("0 pmol", 4)),2)
> > > > > z =  rep(c(rep("Treated", 2), rep("Untreated", 2), rep("Treated", 2),
> > > > > rep("Untreated", 2)))
> > > > > w = c(rep("1:1000", 8), rep("1:2000", 8))
> > > > > y = c(1.18, 1.22, 1.52, 2.37, 0.97, 1.08, 1.17, 1.58, 1.16, 1.16, 2.27, 2.24,
> > > > >       1.26, 1.28, 1.39, 1.77)
> > > > > Q = data.frame(x, z, w, y, stringsAsFactors = FALSE)
> > > > > names(Q) = c("[Target]", "Enzyme", "[Antibody]", "OD")
> > > > > names(Q) = c("x", "z", "w", "y") # for ease, I'll use single letters
> > > > > library(plotrix)
> > > > > brkdn.plot(y~x+z, data=Q,
> > > > >            pch = c(1, 16), cex = 1.5, type="p",
> > > > >            main="Single Measurement",
> > > > >            xlab=expression(bold("S1 nuclease")),
> > > > >            ylab=expression(bold("Optical density")))
> > > > > brkdn.plot(y~x+z+w, data=Q,
> > > > >            pch = c(1, 16), cex = 1.5, type="p",
> > > > >            main="Double Measurement",
> > > > >            xlab=expression(bold("S1 nuclease")),
> > > > >            ylab=expression(bold("Optical density")))
> > > > > ```
> > > > >
> > > > > ______________________________________________
> > > > > [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> > > > > https://stat.ethz.ch/mailman/listinfo/r-help
> > > > > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > > > > and provide commented, minimal, self-contained, reproducible code.
> > >
> > >
> > >
> > > --
> > > Best regards,
> > > Luigi

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

lm_plot.png (25K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: package plotrix: how to account for two two z categories

Luigi
Thanks! looks nicer already.
I am attaching the figure I temporarily made with normal brkdn.plot
but yours has more information.
Best regards
Luigi

On Fri, Sep 25, 2020 at 1:43 PM Jim Lemon <[hidden email]> wrote:

>
> Hi Luigi,
> Here is a quick example of how points would look in the same
> configuration. Perhaps with end caps to show the nesting it may be
> more like what you want.
>
> Jim
>
> On Thu, Sep 24, 2020 at 6:34 PM Jim Lemon <[hidden email]> wrote:
> >
> > Oh, sorry, forgot about the colors. A list beginning with the color
> > for the overall summary, then colors for the first factor and so on.
> > See the help page for examples.
> >
> > Jim
> >
> > On Thu, Sep 24, 2020 at 6:32 PM Jim Lemon <[hidden email]> wrote:
> > >
> > > Hi Luigi,
> > > I thought a lot about that when I was writing the function. The only
> > > way I could think of to show the nesting was dots with horizontal
> > > lines and it looked messy and was quite hard to visualize the nesting.
> > > If you do have any great ideas I always welcome contributions to
> > > plotrix.
> > >
> > > Jim
> > >
> > > On Thu, Sep 24, 2020 at 6:25 PM Luigi Marongiu <[hidden email]> wrote:
> > > >
> > > > Thank you Jim, that is really nice!
> > > > But is there a way to use dots instead of boxes? and how do I control
> > > > the colours?
> > > > Best regards
> > > > Luigi
> > > >
> > > > On Thu, Sep 24, 2020 at 9:29 AM Jim Lemon <[hidden email]> wrote:
> > > > >
> > > > > Hi Luigi,
> > > > > To display a nested breakdown like this I would suggest barNest. This
> > > > > is one way to display the nesting. Note that if you change the order
> > > > > of the factors in the formula you will get a different plot, so think
> > > > > about how you want the summaries nested. Error bars can only be
> > > > > displayed on the final breakdown.
> > > > >
> > > > > # because you have fairly long labels, use a wide plot
> > > > > x11(width=10)
> > > > > barcol<-list("lightgray",c("#ff00ff","#ff99ff"),
> > > > >  c("#00ff00","#66ff66"),c("#aaaa00","#888800"))
> > > > > barNest(y~x+z+w,Q,main="Double Measurement",
> > > > >  ylab="Response",col=barcol,errbars=TRUE)
> > > > >
> > > > > Jim
> > > > >
> > > > > On Wed, Sep 23, 2020 at 10:48 PM Luigi Marongiu
> > > > > <[hidden email]> wrote:
> > > > > >
> > > > > > Hello,
> > > > > > I have an experiment measuring optical density (OD) when comparing
> > > > > > three parameters:
> > > > > > a) concentration of the target
> > > > > > b) concentration of the probe
> > > > > > c) concentration of the reporter antibody.
> > > > > > Using plotrix I can nicely draw the first two into clusters, but I
> > > > > > can't get separation for the third parameter. is there a way in
> > > > > > plotrix to custer data according to two, let's say, z parameters (I
> > > > > > call the second high-level parameter as w)? For instance, two
> > > > > > clusters, each separated into two subclusters. Or is this more a job
> > > > > > for lattice?
> > > > > > Thank you.
> > > > > >
> > > > > > ```
> > > > > > x = rep(c(rep("1000 pmol", 4), rep("0 pmol", 4)),2)
> > > > > > z =  rep(c(rep("Treated", 2), rep("Untreated", 2), rep("Treated", 2),
> > > > > > rep("Untreated", 2)))
> > > > > > w = c(rep("1:1000", 8), rep("1:2000", 8))
> > > > > > y = c(1.18, 1.22, 1.52, 2.37, 0.97, 1.08, 1.17, 1.58, 1.16, 1.16, 2.27, 2.24,
> > > > > >       1.26, 1.28, 1.39, 1.77)
> > > > > > Q = data.frame(x, z, w, y, stringsAsFactors = FALSE)
> > > > > > names(Q) = c("[Target]", "Enzyme", "[Antibody]", "OD")
> > > > > > names(Q) = c("x", "z", "w", "y") # for ease, I'll use single letters
> > > > > > library(plotrix)
> > > > > > brkdn.plot(y~x+z, data=Q,
> > > > > >            pch = c(1, 16), cex = 1.5, type="p",
> > > > > >            main="Single Measurement",
> > > > > >            xlab=expression(bold("S1 nuclease")),
> > > > > >            ylab=expression(bold("Optical density")))
> > > > > > brkdn.plot(y~x+z+w, data=Q,
> > > > > >            pch = c(1, 16), cex = 1.5, type="p",
> > > > > >            main="Double Measurement",
> > > > > >            xlab=expression(bold("S1 nuclease")),
> > > > > >            ylab=expression(bold("Optical density")))
> > > > > > ```
> > > > > >
> > > > > > ______________________________________________
> > > > > > [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> > > > > > https://stat.ethz.ch/mailman/listinfo/r-help
> > > > > > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > > > > > and provide commented, minimal, self-contained, reproducible code.
> > > >
> > > >
> > > >
> > > > --
> > > > Best regards,
> > > > Luigi


--
Best regards,
Luigi

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Rplot.png (13K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: package plotrix: how to account for two two z categories

Jim Lemon-4
Hi Luigi,
Good illustration. Maybe if I can integrate vertical lines to show the
nesting instead of bars it would work better. While it makes the
nesting much clearer, some people may still accuse you of using a bar
plot.

# new functions with the modifications
source("barNest.R")
source("drawNestedSpans.R")
png("lm_plot.png",width=600)
barNest(y~w+z+x,Q,main="Double Measurement",
 ylab="Response",col=barcol,errbars=TRUE)
dev.off()


Jim

On Fri, Sep 25, 2020 at 9:54 PM Luigi Marongiu <[hidden email]> wrote:
>
> Thanks! looks nicer already.
> I am attaching the figure I temporarily made with normal brkdn.plot
> but yours has more information.
> Best regards
> Luigi
>

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

lm_plot.png (26K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: package plotrix: how to account for two two z categories

Jim Lemon-4
Hmmm, maybe without the caps...

Jim

On Sat, Sep 26, 2020 at 8:20 AM Jim Lemon <[hidden email]> wrote:

>
> Hi Luigi,
> Good illustration. Maybe if I can integrate vertical lines to show the
> nesting instead of bars it would work better. While it makes the
> nesting much clearer, some people may still accuse you of using a bar
> plot.
>
> # new functions with the modifications
> source("barNest.R")
> source("drawNestedSpans.R")
> png("lm_plot.png",width=600)
> barNest(y~w+z+x,Q,main="Double Measurement",
>  ylab="Response",col=barcol,errbars=TRUE)
> dev.off()
>
>
> Jim
>
> On Fri, Sep 25, 2020 at 9:54 PM Luigi Marongiu <[hidden email]> wrote:
> >
> > Thanks! looks nicer already.
> > I am attaching the figure I temporarily made with normal brkdn.plot
> > but yours has more information.
> > Best regards
> > Luigi
> >

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

lm_plot.png (25K) Download Attachment