Well I first thought of using the MFA by considering my data as categorical
but indeed the correspondence analysis methods are better suitable for my
data structure,
Thanks again,
P
--
View this message in context:
http://r.789695.n4.nabble.com/dependent-column-s-in-data-frame-tp4685561p46857
That's what I am looking for. Thanks for being so helpful, I appreciate it
very much.
On Sun, Feb 23, 2014 at 10:16 PM, Gabor Grothendieck <
ggrothendi...@gmail.com> wrote:
> On Sun, Feb 23, 2014 at 9:06 PM, C W wrote:
> > Is there a way to rbind this? Do I have to use use a package like 'zoo
I forgot to send this to the list
It may make things easier
Duncan
-Original Message-
From: Duncan Mackay [mailto:dulca...@bigpond.com]
Sent: Monday, 24 February 2014 10:01
To: 'C W'
Subject: RE: [R] predict.lm() does not take ts objects in formula
Hi
You have not told us what newdata
On Sun, Feb 23, 2014 at 9:06 PM, C W wrote:
> Is there a way to rbind this? Do I have to use use a package like 'zoo' and
> merge()?
>
> p <- predict(model, data.frame(t = 1, q = factor(1, 1:4))
> rbind(tsdat, p)
>
If you want to append this to the end of the series then try this:
ts(c(tsdat,
Is there a way to rbind this? Do I have to use use a package like 'zoo'
and merge()?
p <- predict(model, data.frame(t = 1, q = factor(1, 1:4))
rbind(tsdat, p)
On Sun, Feb 23, 2014 at 8:43 PM, C W wrote:
> I guess this is for anyone in the future.
>
> > predict(model, data.frame(t = 1, q = fac
I guess this is for anyone in the future.
> predict(model, data.frame(t = 1, q = factor(1, 1:4))
This would be the answer. Thanks again, Gabor!
Mike
On Sun, Feb 23, 2014 at 8:23 PM, Gabor Grothendieck wrote:
> On Sun, Feb 23, 2014 at 7:40 PM, C W wrote:
> > Gabor,
> > Your response worked p
On Sun, Feb 23, 2014 at 7:40 PM, C W wrote:
> Gabor,
> Your response worked perfectly, list(t = 1, q = factor(1, 1:4) was what I
> couldn't figure out. Thank you.
>
> In predict.lm(model, newdata), 2nd argument MUST be data.frame(). Why does
> list() also work?
>
Yes, it would be better to use
Gabor,
Your response worked perfectly, list(t = 1, q = factor(1, 1:4) was what I
couldn't figure out. Thank you.
In predict.lm(model, newdata), 2nd argument MUST be data.frame(). Why does
list() also work?
Mike
On Sun, Feb 23, 2014 at 7:32 PM, Gabor Grothendieck wrote:
> On Sun, Feb 23, 20
On Sun, Feb 23, 2014 at 7:25 PM, C W wrote:
> Gabor,
> Let me change newdata since it's confusing.
>
> Suppose I want to predict, year 1990, and quarter 2.
>> newdata <- data.frame(c(1990, 1, 0, 0)
>
> Since Q1 is a baseline, we will only see Q2, Q3, Q4. So, 4 parameters in
> total.
>
The formul
Gabor,
Let me change newdata since it's confusing.
Suppose I want to predict, year 1990, and quarter 2.
> newdata <- data.frame(c(1990, 1, 0, 0)
Since Q1 is a baseline, we will only see Q2, Q3, Q4. So, 4 parameters in
total.
Mike
On Sun, Feb 23, 2014 at 7:13 PM, Gabor Grothendieck wrote:
>
On Sun, Feb 23, 2014 at 6:56 PM, C W wrote:
> Gabor,
> I want the new data to be this,
> newdata <- data.frame(c(1, 0, 0, 0))
>
Its not clear what this means. There are two input variables so we
must specify two inputs.
For example, this would get the prediction for t=1 and for level 1 of
q whic
Gabor,
I want the new data to be this,
newdata <- data.frame(c(1, 0, 0, 0))
Here the code with the new data,
> dat <- rnorm(20)
> tsdat <- ts(dat, start=c(1900, 1), freq=4)
> q <- as.factor(rep(1:4, 5))
> t <- 1:20
> lm(tsdat~t+q)
Call:
lm(formula = tsdat ~ t + q)
Coefficients:
(Intercept)
On Sun, Feb 23, 2014 at 6:34 PM, C W wrote:
> Hello,
> I don't know how to use predict.lm() for ts object.
>
> Here's the time series regression.
> y = t + Q1 + Q2 + Q3 + Q4
>
> Here's my R code,
>
>> dat <- rnorm(20)
>> tsdat <- ts(dat, start=c(1900, 1), freq=4)
>> q <- as.factor(rep(1:4, 5))
>>
HI,
For the first question:
you could do,
vec1 <- c(rep(0:4,each=5),rep(0:4,5:1))
as.vector(tapply(vec1,cumsum(abs(c(0,diff(vec1,FUN=paste,collapse=""))
# [1] "0" "1" "2" "3" "4" "0" "" "222" "33"
#[10] "4"
Second question is not clear. May be this link
Hello list,
Does predict.lm() take ts object. I am aware that it requires data to come
in data.frame
Here's the time series regression.
y = t + Q1 + Q2 + Q3 + Q4
Here's my R code,
> dat <- rnorm(20)
> tsdat <- ts(dat, start=c(1900, 1), freq=4)
> q <- as.factor(rep(1:4, 5))
> t <- 1:20
> lm(ts
Hello,
I don't know how to use predict.lm() for ts object.
Here's the time series regression.
y = t + Q1 + Q2 + Q3 + Q4
Here's my R code,
> dat <- rnorm(20)
> tsdat <- ts(dat, start=c(1900, 1), freq=4)
> q <- as.factor(rep(1:4, 5))
> t <- 1:20
> lm(tsdat~t+q)
Call:
lm(formula = tsdat ~ t + q)
I am trying to encrypt postcodes (zipcodes) which, in my datasets, are
strings with 8 characters, including spaces. By referring to other help
files and other R resources I've built some (clumsy looking) code that turns
a postcode into an output I'm happy with. See below.
library("PKI")
#make a
I am trying to encrypt postcodes (zipcodes) which, in my datasets, are
strings with 8 characters, including spaces. By referring to other help
files and other R resources I've built some (clumsy looking) code that turns
a postcode into an output I'm happy with. See below.
library("PKI")
#make a
Thanks, Yihui!! This works!!
Best wishes,
Ranjan
On Sun, 23 Feb 2014 11:31:35 -0600 Yihui Xie wrote:
> Whenever plotmath is unable to provide what you want, you can always
> go to tikzDevice for native LaTeX support:
> http://yihui.name/en/2011/04/produce-authentic-math-formulas-in-r-graphics/
Whenever plotmath is unable to provide what you want, you can always
go to tikzDevice for native LaTeX support:
http://yihui.name/en/2011/04/produce-authentic-math-formulas-in-r-graphics/
Here is an example:
library(tikzDevice)
tikz('test-math.tex', standAlone = TRUE)
plot(1, main = '$\\check{Y}$'
On Feb 23, 2014, at 7:11 AM, Luigi Marongiu wrote:
> Dear all,
Dear Luigi;
Please stop posting triplicate postings and also learn to post in plain text.
>
> I would like to draw a Venn plot for data represented by 6 variables (or
> sets). I know how to do this using the package venneuler (wh
Dear all,
I would like to draw a Venn plot for data represented by 6 variables (or
sets). I know how to do this using the package venneuler (which requires
rJava). However this package does not report the numbers of elements within
each class.
Do you know an alternative to this package to draw th
Dear all,
I would like to draw a Venn plot for data represented by 6 variables (or
sets). I know how to do this using the package venneuler (which requires
rJava). However this package does not report the numbers of elements within
each class.
Do you know an alternative to this package to draw th
Dear all,
I would like to draw a Venn plot for data represented by 6 variables (or
sets). I know how to do this using the package venneuler (which requires
rJava). However this package does not report the numbers of elements within
each class.
Do you know an alternative to this package to draw th
24 matches
Mail list logo