Re: [R] t.test
matthew wrote: Hi, how can I solve a problem without the function t.test??? for example: x-(1,3,5,7) y-(2,4,6) t.test(x,y,alternative=less,paired=FALSE,var.equal=TRUE,conf.level=0.95) Homework? Hints: Take out your statistics textbook and look up the formulas for the two-sample t. You'll probably (there can be some variation depending on the book) find that you need to compute - difference of means - sd for each group - pooled sd - s.e. of differences of means all of which you can do easily in R, once you have the formulas. Then calculate the t statistic and the corresponding p value, either using a table or R's function for the t distibution. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list 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.
Re: [R] t.test
and do read the R-manual about how to make a vector matthew wrote: Hi, how can I solve a problem without the function t.test??? for example: x-(1,3,5,7) y-(2,4,6) t.test(x,y,alternative=less,paired=FALSE,var.equal=TRUE,conf.level=0.95) -- View this message in context: http://www.nabble.com/t.test-tf4034225.html#a11463812 Sent from the R help mailing list archive at Nabble.com. __ R-help@stat.math.ethz.ch mailing list 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.
Re: [R] t.test()
On 23 Nov 2006, at 13:46, Peter Dalgaard wrote:
Robin Hankin [EMAIL PROTECTED] writes:
Hi
I have a vector x of length n. I am interested in x[1]
being different from the other observations (ie x[-1]).
[snip]
What arguments do I need to send to t.test() to test my null?
[snip]
Alternatively, just write up the formula for the t statistic:
x - c(23,25,29,27,30,30)
(x[1]-mean(x[-1]))/sqrt(var(x[-1])*(1+1/(length(x)-1)))
The gotcha in Peter Dalgaard's formula is that the maximum
likelihood estimate for the
variance, with a sample size of one, is zero. This is why var[x[1]]
doesn't appear.
[R reports var(1) as NA because it uses the unbiased formula with
(n-1) on the denominator,
as documented]
Last night I derived the likelihood test for testing my null. Consider
H1: x~N(mu_x,s^2); y~N(mu_y,s^2)
H2: x,y~N(mu,s^2)
The support gained by allowing the two means to differ [ie compare H2
to H1] is:
\[
E=
\frac{n}{2}\ln\left(
\frac{\sum(z_i-\overline{z})^2}{
\sum(x_i-\overline{x})^2+
\sum(y_i-\overline{y})^2
}\right)
\]
where z=c(x,y) is both sets of observations taken together. This
formula supposes one
uses the appropriate maximum likelihood estimate for the (common)
variance. Note that
the MLE for the variance is different on H1 and H2.
Thus if E 2 we can reject H2 and if !(E 2) we can accept (sic) H2.
Robin Hankin
Uncertainty Analyst
National Oceanography Centre, Southampton
European Way, Southampton SO14 3ZH, UK
tel 023-8059-7743
__
R-help@stat.math.ethz.ch mailing list
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.
Re: [R] t.test()
There is no such thing as an unpaired t-test. A t-test can be a paired, one sample or two sample t-test. Since you want to compare the sample against a given mean, you need a one sample t-test. You tried to do a two sample test. That didn't work because you need at least two observations in each group. x - c(23,25,29,27,30,30) t.test(x[-1], mu = x[1]) One Sample t-test data: x[-1] t = 5.3634, df = 4, p-value = 0.005833 alternative hypothesis: true mean is not equal to 23 95 percent confidence interval: 25.50814 30.89186 sample estimates: mean of x 28.2 Cheers, Thierry ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Reseach Institute for Nature and Forest Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, methodology and quality assurance Gaverstraat 4 9500 Geraardsbergen Belgium tel. + 32 54/436 185 [EMAIL PROTECTED] www.inbo.be Do not put your faith in what statistics say until you have carefully considered what they do not say. ~William W. Watt A statistical analysis, properly conducted, is a delicate dissection of uncertainties, a surgery of suppositions. ~M.J.Moroney -Oorspronkelijk bericht- Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Namens Robin Hankin Verzonden: donderdag 23 november 2006 14:12 Aan: [EMAIL PROTECTED] Onderwerp: [R] t.test() Hi I have a vector x of length n. I am interested in x[1] being different from the other observations (ie x[-1]). My null hypothesis is that x[1] is drawn from a Gaussian distribution of the same mean as observations x[-1], which are assumed to be iid Gaussian. The (unknown) variance of x[1] is assumed to be the same as the variance of x[-1]. This should be an unpaired t-test. But x - c(23,25,29,27,30,30) t.test(x=x[1] , y=x[-1]) Error in t.test.default(x = x[1], y = x[-1]) : not enough 'x' observations What arguments do I need to send to t.test() to test my null? -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ R-help@stat.math.ethz.ch mailing list 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@stat.math.ethz.ch mailing list 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.
Re: [R] t.test()
Hello everybody
thanks for your advice here.
I think I'm getting tangled up.
If I use Thierry's test on iid Gaussian data,
the returned p-value should be uniform(0,1),
right?
OK,
R f - function(x){t.test(x=x[-1],mu=x[1])$p.value}
R hist(replicate(1000,f(rnorm(5
This is very skewed towards zero. Why isn't the
histogram showing a uniform distribution?
On 23 Nov 2006, at 13:32, ONKELINX, Thierry wrote:
There is no such thing as an unpaired t-test. A t-test can be a
paired,
one sample or two sample t-test. Since you want to compare the sample
against a given mean, you need a one sample t-test. You tried to do a
two sample test. That didn't work because you need at least two
observations in each group.
x - c(23,25,29,27,30,30)
t.test(x[-1], mu = x[1])
One Sample t-test
data: x[-1]
t = 5.3634, df = 4, p-value = 0.005833
alternative hypothesis: true mean is not equal to 23
95 percent confidence interval:
25.50814 30.89186
sample estimates:
mean of x
28.2
Cheers,
Thierry
--
--
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Reseach Institute for Nature
and Forest
Cel biometrie, methodologie en kwaliteitszorg / Section biometrics,
methodology and quality assurance
Gaverstraat 4
9500 Geraardsbergen
Belgium
tel. + 32 54/436 185
[EMAIL PROTECTED]
www.inbo.be
Do not put your faith in what statistics say until you have carefully
considered what they do not say. ~William W. Watt
A statistical analysis, properly conducted, is a delicate
dissection of
uncertainties, a surgery of suppositions. ~M.J.Moroney
-Oorspronkelijk bericht-
Van: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Namens Robin Hankin
Verzonden: donderdag 23 november 2006 14:12
Aan: [EMAIL PROTECTED]
Onderwerp: [R] t.test()
Hi
I have a vector x of length n. I am interested in x[1]
being different from the other observations (ie x[-1]).
My null hypothesis is that x[1]
is drawn from a Gaussian distribution of the same
mean as observations x[-1], which are assumed
to be iid Gaussian. The (unknown) variance
of x[1] is assumed to be the same as the
variance of x[-1].
This should be an unpaired t-test.
But
x - c(23,25,29,27,30,30)
t.test(x=x[1] , y=x[-1])
Error in t.test.default(x = x[1], y = x[-1]) :
not enough 'x' observations
What arguments do I need to send to t.test() to test my null?
--
Robin Hankin
Uncertainty Analyst
National Oceanography Centre, Southampton
European Way, Southampton SO14 3ZH, UK
tel 023-8059-7743
__
R-help@stat.math.ethz.ch mailing list
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.
--
Robin Hankin
Uncertainty Analyst
National Oceanography Centre, Southampton
European Way, Southampton SO14 3ZH, UK
tel 023-8059-7743
__
R-help@stat.math.ethz.ch mailing list
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.
Re: [R] t.test()
ONKELINX, Thierry [EMAIL PROTECTED] writes: There is no such thing as an unpaired t-test. A t-test can be a paired, one sample or two sample t-test. Since you want to compare the sample against a given mean, you need a one sample t-test. You tried to do a two sample test. That didn't work because you need at least two observations in each group. NO! This is just plain wrong. x - c(23,25,29,27,30,30) t.test(x[-1], mu = x[1]) One Sample t-test data: x[-1] t = 5.3634, df = 4, p-value = 0.005833 alternative hypothesis: true mean is not equal to 23 95 percent confidence interval: 25.50814 30.89186 sample estimates: mean of x 28.2 Cheers, Thierry ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Reseach Institute for Nature and Forest Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, methodology and quality assurance Gaverstraat 4 9500 Geraardsbergen Belgium tel. + 32 54/436 185 [EMAIL PROTECTED] www.inbo.be Do not put your faith in what statistics say until you have carefully considered what they do not say. ~William W. Watt A statistical analysis, properly conducted, is a delicate dissection of uncertainties, a surgery of suppositions. ~M.J.Moroney -Oorspronkelijk bericht- Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Namens Robin Hankin Verzonden: donderdag 23 november 2006 14:12 Aan: [EMAIL PROTECTED] Onderwerp: [R] t.test() Hi I have a vector x of length n. I am interested in x[1] being different from the other observations (ie x[-1]). My null hypothesis is that x[1] is drawn from a Gaussian distribution of the same mean as observations x[-1], which are assumed to be iid Gaussian. The (unknown) variance of x[1] is assumed to be the same as the variance of x[-1]. This should be an unpaired t-test. But x - c(23,25,29,27,30,30) t.test(x=x[1] , y=x[-1]) Error in t.test.default(x = x[1], y = x[-1]) : not enough 'x' observations What arguments do I need to send to t.test() to test my null? -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ R-help@stat.math.ethz.ch mailing list 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@stat.math.ethz.ch mailing list 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. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list 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.
Re: [R] t.test()
Robin Hankin [EMAIL PROTECTED] writes: Hi I have a vector x of length n. I am interested in x[1] being different from the other observations (ie x[-1]). My null hypothesis is that x[1] is drawn from a Gaussian distribution of the same mean as observations x[-1], which are assumed to be iid Gaussian. The (unknown) variance of x[1] is assumed to be the same as the variance of x[-1]. This should be an unpaired t-test. But x - c(23,25,29,27,30,30) t.test(x=x[1] , y=x[-1]) Error in t.test.default(x = x[1], y = x[-1]) : not enough 'x' observations What arguments do I need to send to t.test() to test my null? You can't. Shouldn't be too much of a problem to modify t.test.default to stop it complaining. (It's not quite enough to remove the check for nx 2, though. You also need to deal with var(x) being NA if x has length one.) Alternatively, just write up the formula for the t statistic: x - c(23,25,29,27,30,30) (x[1]-mean(x[-1]))/sqrt(var(x[-1])*(1+1/(length(x)-1))) [1] -2.189595 2*pt(-2.1896,4) [1] 0.09373392 -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list 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.
Re: [R] t.test()
On Thu, 23 Nov 2006, ONKELINX, Thierry wrote: There is no such thing as an unpaired t-test. A t-test can be a paired, one sample or two sample t-test. Since you want to compare the sample against a given mean, you need a one sample t-test. You tried to do a two sample test. That didn't work because you need at least two observations in each group. x - c(23,25,29,27,30,30) t.test(x[-1], mu = x[1]) One Sample t-test data: x[-1] t = 5.3634, df = 4, p-value = 0.005833 alternative hypothesis: true mean is not equal to 23 95 percent confidence interval: 25.50814 30.89186 sample estimates: mean of x 28.2 That is not what Robin is asking for. He has an unpaired two-sample t-test in the terminology I understand, but is also assuming var.equal=TRUE, which is not the default in R and so needs to specified. It should be possible to do x - c(23,25,29,27,30,30) t.test(x=x[1], y=x[-1], var.equal=TRUE) since you can make use of the pooled variance from the two groups. In fact, R does not currently allow that case. -Oorspronkelijk bericht- Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Namens Robin Hankin Verzonden: donderdag 23 november 2006 14:12 Aan: [EMAIL PROTECTED] Onderwerp: [R] t.test() Hi I have a vector x of length n. I am interested in x[1] being different from the other observations (ie x[-1]). My null hypothesis is that x[1] is drawn from a Gaussian distribution of the same mean as observations x[-1], which are assumed to be iid Gaussian. The (unknown) variance of x[1] is assumed to be the same as the variance of x[-1]. This should be an unpaired t-test. But x - c(23,25,29,27,30,30) t.test(x=x[1] , y=x[-1]) Error in t.test.default(x = x[1], y = x[-1]) : not enough 'x' observations What arguments do I need to send to t.test() to test my null? -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ R-help@stat.math.ethz.ch mailing list 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@stat.math.ethz.ch mailing list 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. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list 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.
Re: [R] t.test()
On Thu, 23 Nov 2006, Peter Dalgaard wrote: Robin Hankin [EMAIL PROTECTED] writes: Hi I have a vector x of length n. I am interested in x[1] being different from the other observations (ie x[-1]). My null hypothesis is that x[1] is drawn from a Gaussian distribution of the same mean as observations x[-1], which are assumed to be iid Gaussian. The (unknown) variance of x[1] is assumed to be the same as the variance of x[-1]. This should be an unpaired t-test. But x - c(23,25,29,27,30,30) t.test(x=x[1] , y=x[-1]) Error in t.test.default(x = x[1], y = x[-1]) : not enough 'x' observations What arguments do I need to send to t.test() to test my null? You can't. Shouldn't be too much of a problem to modify t.test.default to stop it complaining. (It's not quite enough to remove the check for nx 2, though. You also need to deal with var(x) being NA if x has length one.) And the check is correct if var.equal=FALSE, the default. I've got a modification for the case var.equal=TRUE running, and will patch R shortly. Alternatively, just write up the formula for the t statistic: x - c(23,25,29,27,30,30) (x[1]-mean(x[-1]))/sqrt(var(x[-1])*(1+1/(length(x)-1))) [1] -2.189595 2*pt(-2.1896,4) [1] 0.09373392 -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list 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.
Re: [R] t.test()
23 Nov 2006 14:48:31 +0100, Peter Dalgaard [EMAIL PROTECTED]: ONKELINX, Thierry [EMAIL PROTECTED] writes: There is no such thing as an unpaired t-test. A t-test can be a paired, one sample or two sample t-test. Since you want to compare the sample against a given mean, you need a one sample t-test. You tried to do a two sample test. That didn't work because you need at least two observations in each group. NO! This is just plain wrong. I agree. Hoping I'm not missing anything, something like: t.test(x[1],x[-1], var.equal=TRUE) should work, since pooled variance can be computed if length(x)2. But, surprising, this still gives not enough 'x' observations. I've seen that's because length of the first argument in t.test is checked before checking 'var.equal' value in 't.test.default'. I'm wrong, or 2 sided test should be computed as: t.test.value - abs(x[1]-mean(x[-1]))/var(x) t.test.pvalue - 2*pt(-t.test.value, length(x)-2) ? x - c(23,25,29,27,30,30) t.test(x[-1], mu = x[1]) One Sample t-test data: x[-1] t = 5.3634, df = 4, p-value = 0.005833 alternative hypothesis: true mean is not equal to 23 95 percent confidence interval: 25.50814 30.89186 sample estimates: mean of x 28.2 Cheers, Thierry ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Reseach Institute for Nature and Forest Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, methodology and quality assurance Gaverstraat 4 9500 Geraardsbergen Belgium tel. + 32 54/436 185 [EMAIL PROTECTED] www.inbo.be Do not put your faith in what statistics say until you have carefully considered what they do not say. ~William W. Watt A statistical analysis, properly conducted, is a delicate dissection of uncertainties, a surgery of suppositions. ~M.J.Moroney -Oorspronkelijk bericht- Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Namens Robin Hankin Verzonden: donderdag 23 november 2006 14:12 Aan: [EMAIL PROTECTED] Onderwerp: [R] t.test() Hi I have a vector x of length n. I am interested in x[1] being different from the other observations (ie x[-1]). My null hypothesis is that x[1] is drawn from a Gaussian distribution of the same mean as observations x[-1], which are assumed to be iid Gaussian. The (unknown) variance of x[1] is assumed to be the same as the variance of x[-1]. This should be an unpaired t-test. But x - c(23,25,29,27,30,30) t.test(x=x[1] , y=x[-1]) Error in t.test.default(x = x[1], y = x[-1]) : not enough 'x' observations What arguments do I need to send to t.test() to test my null? -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ R-help@stat.math.ethz.ch mailing list 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@stat.math.ethz.ch mailing list 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. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list 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@stat.math.ethz.ch mailing list 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.
Re: [R] t.test()
Hi, seems as if you want to test if a single subject is a member of a population from which you have drawn a sample. This question has been addressed by Payne Jones and Crawford and collaborators. You should have a look at their articles in case that I interpreted your null-hypothesis and data in the correct way Good luck, Thilo Payne Jones (1957, J Clin Psychol 13:115-121) Crawford, Howell, Garthwaite (1998, J Clin Exp Neurosychol 12:482) Crawford JR, Garthwaite PH (2005) Evaluation of Criteria for Classical Dissociations in Single-Case Studies by Monte Carlo Simulation. Neuropsychology 19(5): 664-678) On Thursday 23 November 2006 14:48, Peter Dalgaard wrote: ONKELINX, Thierry [EMAIL PROTECTED] writes: There is no such thing as an unpaired t-test. A t-test can be a paired, one sample or two sample t-test. Since you want to compare the sample against a given mean, you need a one sample t-test. You tried to do a two sample test. That didn't work because you need at least two observations in each group. NO! This is just plain wrong. x - c(23,25,29,27,30,30) t.test(x[-1], mu = x[1]) One Sample t-test data: x[-1] t = 5.3634, df = 4, p-value = 0.005833 alternative hypothesis: true mean is not equal to 23 95 percent confidence interval: 25.50814 30.89186 sample estimates: mean of x 28.2 Cheers, Thierry ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Reseach Institute for Nature and Forest Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, methodology and quality assurance Gaverstraat 4 9500 Geraardsbergen Belgium tel. + 32 54/436 185 [EMAIL PROTECTED] www.inbo.be Do not put your faith in what statistics say until you have carefully considered what they do not say. ~William W. Watt A statistical analysis, properly conducted, is a delicate dissection of uncertainties, a surgery of suppositions. ~M.J.Moroney -Oorspronkelijk bericht- Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Namens Robin Hankin Verzonden: donderdag 23 november 2006 14:12 Aan: [EMAIL PROTECTED] Onderwerp: [R] t.test() Hi I have a vector x of length n. I am interested in x[1] being different from the other observations (ie x[-1]). My null hypothesis is that x[1] is drawn from a Gaussian distribution of the same mean as observations x[-1], which are assumed to be iid Gaussian. The (unknown) variance of x[1] is assumed to be the same as the variance of x[-1]. This should be an unpaired t-test. But x - c(23,25,29,27,30,30) t.test(x=x[1] , y=x[-1]) Error in t.test.default(x = x[1], y = x[-1]) : not enough 'x' observations What arguments do I need to send to t.test() to test my null? -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ R-help@stat.math.ethz.ch mailing list 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@stat.math.ethz.ch mailing list 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. -- Thilo Kellermann Department of Psychiatry und Psychotherapy RWTH Aachen University Pauwelstr. 30 52074 Aachen Tel.: +49 (0)241 / 8089977 Fax.: +49 (0)241 / 8082401 E-Mail: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list 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.
Re: [R] t.test question
Roth, Richard wrote: Hi, I have a data matrix of gene expression data from two groups that I would like to compare using the t-test. The data has been processed using RMA and transformed using log2. I would like to compare the two groups for each gene (N=10,000 genes) and have a result that lists the p-value for each test. Can anyone help me with the t.test setup for this analysis? Take a look at the limma package in BioConductor. You will be able to do what you want much faster than using e.g., t.test(). There is an extensive user's guide that should help you figure out what to do. HTH, Jim Thanks, Rich Rich Roth, PhD Senior Scientist Molecular Medicine Neurocrine Biosciences 858-617-7204 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- James W. MacDonald, M.S. Biostatistician Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] t.test using RSPerl
This has nothing to do with RSPerl, instead it has to do what kind of object you obtain and how these are print():ed. Typing the name of an object, say, 'res', at R prompt and pressing ENTER; res is equivalent as typing print(res) This is for convenience to the user. Basically, this is why you can do 1+1 [1] 2 without having to do print(1+1) [1] 2 When you transfer the object from R to Perl you will, as expect, only receive the value of 'res', not the output from 'print(res)'. Here is an example illustrating the behavior in R: res - t.test(1:10,y=c(7:20)) length(res) [1] 9 str(res) List of 9 $ statistic : Named num -5.43 ..- attr(*, names)= chr t $ parameter : Named num 22 ..- attr(*, names)= chr df $ p.value: num 1.86e-05 $ conf.int : atomic [1:2] -11.05 -4.95 ..- attr(*, conf.level)= num 0.95 $ estimate : Named num [1:2] 5.5 13.5 ..- attr(*, names)= chr [1:2] mean of x mean of y $ null.value : Named num 0 ..- attr(*, names)= chr difference in means $ alternative: chr two.sided $ method : chr Welch Two Sample t-test $ data.name : chr 1:10 and c(7:20) - attr(*, class)= chr htest print(res) Welch Two Sample t-test data: 1:10 and c(7:20) t = -5.4349, df = 21.982, p-value = 1.855e-05 alternative hypothesis: true difference in means is not equal to 0 95 percent confidence interval: -11.052802 -4.947198 sample estimates: mean of x mean of y 5.5 13.5 In your case, to capture what print(res) is outputting, you may want to look at ?capture.out, that is output - capture.output(res) output # ...that is, print(output) [1] [2] \tWelch Two Sample t-test [3] [4] data: 1:10 and c(7:20) [5] t = -5.4349, df = 21.982, p-value = 1.855e-05 [6] alternative hypothesis: true difference in means is not equal to 0 [7] 95 percent confidence interval: [8] -11.052802 -4.947198 [9] sample estimates: [10] mean of x mean of y [11] 5.5 13.5 [12] and transfer 'output' to Perl. Cheers Henrik Wagle, Mugdha wrote: Hi, I've just started using R and RSPerl. I have some code as follows: R::initR(--no-save); R::call(t.test, ([EMAIL PROTECTED], [EMAIL PROTECTED])); where @array1 and @array2 are both 1-dimensional arrays in Perl having 54675 elements each. On execution the output is as follows: Calling R function name `t.test', # arguments: 3 1) Arg type 3 Got a reference to a value 10 Here now!2) Arg type 3 Got a reference to a value 10 Here now!Calling R t.test(c(0, 6.24280675278087, 6.35175793656943, 5.76925805661511, 7.0789316246711, 7.4636498661157, 8.13730810691084, 8.78203131644273, 9.64502765609435, 9.95631242346133, 5.83129579495516, 6.8798700754926, 7.31814159140937...(REST OF THE ARRAY ELEMENTS). 4.91632461462501, 3.38099467434464, 3.91800507710569, 3.23867845216438, 3.38439026334577, 4.64918707140487, 3.23474917402449, 3.62966009445396, 3.36729582998647, 3.91999117507732 )) Performed the call, result has length 9 My question is : with other functions such as sum and log10, the actual values of the result are displayed. Here the call seems to have worked but the output is not what you get when running t.test directly on the R command prompt.. data: data4[2] and data4[3] t = 0.2186, df = 109.847, p-value = 0.8274 alternative hypothesis: true difference in means is not equal to 0 95 percent confidence interval: -3722.830 4645.723 sample estimates: mean of x mean of y 6185.139 5723.693 which is what I had expected, after seeing the outputs in the case of simpler functions like sum. Could anyone please tell me how I can obtain the output I expect(i.e. the same as the command line outputgiving values of t, p-value and the means)? Thank you very much for the help!! Sincerely, Mugdha Wagle Hartwell Center for Bioinformatics and Biotechnology, St.Jude Children's Research Hospital, Memphis TN 38105 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] t.test formatting question
as.numeric() (and its siblings) strip the names from vectors, e.g.: as.numeric(t.test(rnorm(1001))$statistic) [1] -0.6320304 hth, Tony Plate At Friday 05:14 PM 4/16/2004, christopher ciotti wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hello all - I'm trying to format some data where I only need one of the values returned from a test, say a t-test in this instance. I have the following: R.version.string [1] R version 1.9.0, 2004-04-12 x - rnorm(1001) t.test(x)$statistic t 0.321 Any easy way (other than straight calculation) to get the t-statistic w/o the 't' above it? TIA - -- chris ciotti ([EMAIL PROTECTED]) PGP ID: 0xE94BB3B7 -BEGIN PGP SIGNATURE- Version: PGP 8.0 iQA/AwUBQIBotlkgIqbpS7O3EQIt7ACdEhhoM3tqqmDae71Rdo/eHg2TvoQAoLSw 8QjwkHENOjjs5mzCWlXSTQXO =LvW5 -END PGP SIGNATURE- __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] t.test formatting question
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Tony Plate wrote: as.numeric() (and its siblings) strip the names from vectors, e.g.: as.numeric(t.test(rnorm(1001))$statistic) [1] -0.6320304 hth, Tony Plate Thanks for the quick response. - -- chris ciotti ([EMAIL PROTECTED]) PGP ID: 0xE94BB3B7 -BEGIN PGP SIGNATURE- Version: PGP 8.0 iQA/AwUBQIBqXlkgIqbpS7O3EQKNRwCfREBcPiBQDxIKcnUD6uVJyt5YP/wAnjBR l+C3+RSz2r0juvBZ87DTyyRI =S8Mz -END PGP SIGNATURE- __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] t.test formatting question
How about this? t.test(x)[[1]] The result of t.test(x) is a list and statistic is the first component of that list. christopher ciotti wrote: ... I'm trying to format some data where I only need one of the values returned from a test, say a t-test in this instance. I have the following: R.version.string [1] R version 1.9.0, 2004-04-12 x - rnorm(1001) t.test(x)$statistic t 0.321 Any easy way (other than straight calculation) to get the t-statistic w/o the 't' above it? TIA ... -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 452-1424 (M, W, F) fax: (917) 438-0894 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] t.test formatting question
Sorry, I should have checked that more closely. I see that t.test(rnorm(12))[[1]] retains the t. christopher ciotti wrote: ... I'm trying to format some data where I only need one of the values returned from a test, say a t-test in this instance. I have the following: R.version.string [1] R version 1.9.0, 2004-04-12 x - rnorm(1001) t.test(x)$statistic t 0.321 Any easy way (other than straight calculation) to get the t-statistic w/o the 't' above it? TIA ... -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 452-1424 (M, W, F) fax: (917) 438-0894 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] t.test formatting question
How about: as.numeric(t.test(rnorm(12))[[1]]) - Original Message - From: Chuck Cleland [EMAIL PROTECTED] To: christopher ciotti [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Sent: Friday, April 16, 2004 6:25 PM Subject: Re: [R] t.test formatting question Sorry, I should have checked that more closely. I see that t.test(rnorm(12))[[1]] retains the t. christopher ciotti wrote: ... I'm trying to format some data where I only need one of the values returned from a test, say a t-test in this instance. I have the following: R.version.string [1] R version 1.9.0, 2004-04-12 x - rnorm(1001) t.test(x)$statistic t 0.321 Any easy way (other than straight calculation) to get the t-statistic w/o the 't' above it? TIA ... -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 452-1424 (M, W, F) fax: (917) 438-0894 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
