R help - Nov 2017 - Scatterplot of many variables against a single variable

Dear

I try to realize one scatter matrix which draws *one single variable to all
variables* with *regression line* . You can see my eviews version  in the
annex .

How can I draw this graph with R studio?


Sincerely
Engin YILMAZ

Hi Engin,
Sadly, your illustration was ambushed on the way to the list. Perhaps
you want something like this:

# proportion of useful answers to your request
pua<-sort(runif(20))
#legibility of your request
lor<-sort(runif(20))+runif(20,-0.5,0.5)
# is a data set provided?
dsp<-sort(runif(20))+runif(20,-0.5,0.5)
# generate a linear model for the above
pua.lm<-lm(pua~lor+dsp)
# get the coefficients
pua.lm

Call:
lm(formula = pua ~ lor + dsp)

Coefficients:
(Intercept)          lor          dsp
    0.1692       0.6132       0.3311

plot(pua~lor,col="red",main="Proportion of useful answers by
request quality")
points(pua~dsp,col="blue",pch=2)
abline(0.1692,0.6132,col="red")
abline(0.1692,0.3311,col="blue")

So, the more readable your request and the quality of the data that
you provide, the more useful answers you are likely to receive.

Jim


On Mon, Nov 27, 2017 at 7:56 PM, Engin YILMAZ <ispanyolcom at gmail.com>
wrote:
> Dear
>
> I try to realize one scatter matrix which draws *one single variable to all
> variables* with *regression line* . You can see my eviews version  in the
> annex .
>
> How can I draw this graph with R studio?
>
>
> Sincerely
> Engin YILMAZ
> ______________________________________________
> R-help at r-project.org 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.

LOL. Great reply Jim.
(N.B. Jim's conclusion is "debatable" by a judicious choice of
seed. e.g.
set.seed(79) suggests that making the request more readable will actually
lower the number of useful answers. :-))


On Mon, Nov 27, 2017 at 11:42 AM, Jim Lemon <drjimlemon at gmail.com>
wrote:
> Hi Engin,
> Sadly, your illustration was ambushed on the way to the list. Perhaps
> you want something like this:
>
> # proportion of useful answers to your request
> pua<-sort(runif(20))
> #legibility of your request
> lor<-sort(runif(20))+runif(20,-0.5,0.5)
> # is a data set provided?
> dsp<-sort(runif(20))+runif(20,-0.5,0.5)
> # generate a linear model for the above
> pua.lm<-lm(pua~lor+dsp)
> # get the coefficients
> pua.lm
>
> Call:
> lm(formula = pua ~ lor + dsp)
>
> Coefficients:
> (Intercept)          lor          dsp
>     0.1692       0.6132       0.3311
>
> plot(pua~lor,col="red",main="Proportion of useful answers by
request
> quality")
> points(pua~dsp,col="blue",pch=2)
> abline(0.1692,0.6132,col="red")
> abline(0.1692,0.3311,col="blue")
>
> So, the more readable your request and the quality of the data that
> you provide, the more useful answers you are likely to receive.
>
> Jim
>
>
> On Mon, Nov 27, 2017 at 7:56 PM, Engin YILMAZ <ispanyolcom at
gmail.com>
> wrote:
> > Dear
> >
> > I try to realize one scatter matrix which draws *one single variable
to
> all
> > variables* with *regression line* . You can see my eviews version  in
the
> > annex .
> >
> > How can I draw this graph with R studio?
> >
> >
> > Sincerely
> > Engin YILMAZ
> > ______________________________________________
> > R-help at r-project.org 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.
>
> ______________________________________________
> R-help at r-project.org 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.
>
	[[alternative HTML version deleted]]

Here's the quickest way I know of to get a scatterplot of many variables
against a single variable. I create example data to illustrate.

x <- 1:10
ys <- matrix( runif(30), ncol=3)

matplot(x,ys)
## or, a little better,
matplot(x,ys, type='b')

To add regression lines:

for (iy in seq(ncol(ys))) abline(lsfit(x, ys[,iy]))

Coloring the regression lines to match the matplot will take a little more work
(but not a lot more).

Don't ask me about putting the regression line formulas on the plot like in
Excel. I don't do things that way...

There is probably something in ggplot2 or some other package.

-Don

--
Don MacQueen
Lawrence Livermore National Laboratory
7000 East Ave., L-627
Livermore, CA 94550
925-423-1062
Lab cell 925-724-7509
 
 

On 11/27/17, 12:56 AM, "R-help on behalf of Engin YILMAZ"
<r-help-bounces at r-project.org on behalf of ispanyolcom at gmail.com>
wrote:

    Dear
    
    I try to realize one scatter matrix which draws *one single variable to all
    variables* with *regression line* . You can see my eviews version  in the
    annex .
    
    How can I draw this graph with R studio?
    
    
    Sincerely
    Engin YILMAZ
    ______________________________________________
    R-help at r-project.org 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.

> On 27 Nov 2017, at 11:56, Engin YILMAZ <ispanyolcom at gmail.com>
wrote:
> 
> Dear
> 
> I try to realize one scatter matrix which draws *one single variable to all
> variables* with *regression line* . You can see my eviews version  in the
> annex .
> 
> How can I draw this graph with R studio?
A tiny note; You do calculations in R not RSudio. RStudio is a tool (IDE) to use
R in an easy way.

The code below shows how to accomplish this task easily by ggplot. It?s adapted
from [1].

library(ggplot2)
library(reshape2)

# This is your initial data.frame and you want scatterplots of all variables
against x1.
foo <- data.frame(x1 = runif(50, 0, 1), 
                  x2 = runif(50, 0, 1), 
                  x3 = runif(50, 0, 1), 
                  x4 = runif(50, 0, 1))

# melt data. This is very handy function from reshape2 library.
foo2 <- melt(foo, "x1?)

# plot points and add lm lines.
ggplot(foo2, aes(value, x1)) +  
  geom_point() + 
  geom_smooth(method=lm) +
  facet_grid(.~variable)


1-
https://stackoverflow.com/questions/24648729/plot-one-numeric-variable-against-n-numeric-variables-in-n-plots

isezen

Thanks all users

<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free.
www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

2017-11-27 23:18 GMT+03:00 Ismail SEZEN <sezenismail at gmail.com>:
>
> > On 27 Nov 2017, at 11:56, Engin YILMAZ <ispanyolcom at
gmail.com> wrote:
> >
> > Dear
> >
> > I try to realize one scatter matrix which draws *one single variable
to
> all
> > variables* with *regression line* . You can see my eviews version  in
the
> > annex .
> >
> > How can I draw this graph with R studio?
>
> A tiny note; You do calculations in R not RSudio. RStudio is a tool (IDE)
> to use R in an easy way.
>
> The code below shows how to accomplish this task easily by ggplot. It?s
> adapted from [1].
>
> library(ggplot2)
> library(reshape2)
>
> # This is your initial data.frame and you want scatterplots of all
> variables against x1.
> foo <- data.frame(x1 = runif(50, 0, 1),
>                   x2 = runif(50, 0, 1),
>                   x3 = runif(50, 0, 1),
>                   x4 = runif(50, 0, 1))
>
> # melt data. This is very handy function from reshape2 library.
> foo2 <- melt(foo, "x1?)
>
> # plot points and add lm lines.
> ggplot(foo2, aes(value, x1)) +
>   geom_point() +
>   geom_smooth(method=lm) +
>   facet_grid(.~variable)
>
>
> 1- https://stackoverflow.com/questions/24648729/plot-one-
> numeric-variable-against-n-numeric-variables-in-n-plots
>
> isezen



-- 
*Sayg?lar?mla*
Engin YILMAZ

	[[alternative HTML version deleted]]