Re: [R] New var

Ii is difficult to provide useful help, because you have failed to read and follow the posting guide. In particular: 1. Plain text, not HTML. 2. Use dput() or provide code to create your example. Text printouts such as that which you gave require some work to wrangle into into an example that we

Re: [R] New var

You do understand that this is not the "do my work for me" mailing list, don't you? You should be asking questions like "why doesn't my code do one of these if conditions?" so that you know HOW to write the rest of them yourself. In addition you have been posting here long enough to know that

[R] New var

Hi all, I have a data set with time interval and depending on the interval I want to create 5 more variables . Sample data below obs, Start, End 1,2/1/2015, 1/1/2017 2,4/11/2010, 1/1/2011 3,1/4/2006, 5/3/2007 4,10/1/2007, 1/1/2008 5,6/1/2011, 1/1/2012 6,10/15/2004,12/1/2004 First, I want

Re: [R] comparing columns and printing overlapping rows

There are missing details, such as: what do you mean by "overlapping"? do the "files" have the same number of rows? do you care whether the "overlapping" entries are in the same row? what kind of R data structure do you have the "files" stored in? Assuming your file 1 is stored in a

[R] comparing columns and printing overlapping rows

Hi All, I have two files. 1. with only one column 2. data matrix I need to compare first columns of both files and print the rows from second file for the overlapping entries. I have solutions for awk and sed, but I need how to do it in R. Thanks Regards Anchal -- Anchal Sharma, PhD

Re: [R] subletting an array according to dimnames

That works. Thank you! 2017-06-02 8:00 GMT-04:00 Adams, Jean : > Have you tried P2["20", "10", "0"] ? > > Jean > > On Thu, Jun 1, 2017 at 3:10 PM, li li wrote: > >> Hi all, >> I have a three dimensional array with the corresponding dimension names. >> I

Re: [R] Reversing one dimension of an array, in a generalized case

At the very real risk of flogging a dead horse, I realized that it is slightly cleaner to use the logical index TRUE in the argument list of `[` rather than the seq_len(dim[i]) to indicate "everything" in the ith array dimension. I post on list because maybe this might be a bit informative to

Re: [R] Problem of a function I wrote

Hi only some attachment types are alowed, see Posting Guide. Post your code with some toy data directly into body of your mail text. From your explanation I wonder why you append vaues to vector, it is rarely necessary. Cheers Petr > -Original Message- > From: R-help

Re: [R] Upper bands and lower bands

Hi seems to me like homework, this list has no homework policy. For linear decay you can use > temp<-c(10,9,8,7,6,5,4,3,2,1) > plus <- seq(10,5, length=10)/100 > plot(1:10, temp, ylim=c(0, max(temp+plus))) > lines(1:10, temp+temp*plus) > lines(1:10, temp-temp*plus) > for exponential you should

[R] R 3.4.1 at end of June

Just a quick note to say that we intend to have a patch release, probably on June 30, mainly to pick up a few balls that were dropped on the 3.4.0 release. Full schedule to appear later (just need a little time to actually set it up + checking that the date doesn't collide with schedules of

Re: [R] modEvA D-squared for gamma glm

... As this concerns gene expression, perhaps the Bioconductor list might be more appropriate. As it is prmarily statistics, not R programming, maybe stats.stackexchange.com might be a better venue. You might also be better off contacting the package maintainers (?maintainer) for such

[R] modEvA D-squared for gamma glm

Hi All, I am running a generalized linear model with gamma distribution in R (glm, family=gamma ) for my data (gene expression as response variable and few predictors). I want to calculate r-squared for this model. I have been reading online about it and found there are multiple formulas for

Re: [R-es] CV en R

No es que me parezca mal, pero es mucho mejor, como te hemos recomendado, que uses "caret". Te simplificar� mucho la vida, y te proporcionar� visualizaciones de los resultados de la validaci�n que est�n muy bien. En fin, suerte. De: Jes�s Para Fern�ndez

Re: [R-es] CV en R

No, llega un momento en el que m�s �rboles no te supone mejor�a, e incluso funciona peor. Que funcione peor lo atribuyo al ruido, porque en teor�a no tiene mucho sentido, la verdad... Pero no he probado a coger m�s �rboles de los "necesarios". Lo probar�� Un saludo De: Jes�s Para Fern�ndez

Re: [R-es] CV en R

Si te fijas, los mismos autores te indican que mires qué parámetros se ajustan mejor: https://www.stat.berkeley.edu/~breiman/RandomForests/cc_manual.htm#l2 Un saludo Isidro Hidalgo Arellano Observatorio del Mercado de Trabajo Consejería de Economía, Empresas y Empleo

Re: [R-es] CV en R

Hola, No soy un experto en estas técnicas, pero hasta donde yo se, el algoritmo Random Forest no requiere cross validation. Lo dice el mismo Leo Breiman que creo ha sido uno de los investigadores que más ha contribuido al desarrollo de Random Forest (

Re: [R] subletting an array according to dimnames

Have you tried P2["20", "10", "0"] ? Jean On Thu, Jun 1, 2017 at 3:10 PM, li li wrote: > Hi all, > I have a three dimensional array with the corresponding dimension names. > I would like to subset the array according to the dimension names. For > example, > suppose I

Re: [R] Upper bands and lower bands

Hi Pedro, If I get the idea, you want a vector of length 7300 starting at 1 and ending at 0. seq(1,0,length.out=7300) Produces such a vector. It seems to me that: seq(1.1,0.05,length.out=7300) produces the upper vector and: seq(0.95,-0.05 length.out=7300) the lower. Then again, I might have

Re: [R-es] CV en R

Hola, Eso es justamente lo que hace "caret" de una manera muy sencilla y sin que tú te tengas que preocupar de quedarte con el mejor bucket (del CV) o con la mejor combinación en tu "grid search". Te recomiendo que uses "caret" para esto Puedes incluso evaluar los dos algoritmos "RF" y

Re: [R-es] CV en R

No me has parecido para nada borde. Ok. Centr�monos en RF y bajemos el n� de par�metros a 2: ntree y nodesize. Te haces una parrilla de ntree: 100, 200, 300, 400, 500 Otra de nodesize: 3, 6, 10 Con esto tienes 15 combinaciones. Vamos al c�digo. Simplemente crea una lista donde metes los

Re: [R-es] CV en R

No me hab�a fijado en el c�digo, te hab�a he contestado te�ricamente. A ver, en ese c�digo tienes varios problemas: - No especificas los par�metros del modelo (para eso es la validaci�n cruzada). En RF tendr�as que especificar el n�mero de �rboles, la cantidad de puntos con los que

Re: [R-es] CV en R

Abundando en mi respuesta anterior, ni siquiera la validación cruzada es una medida perfecta del comportamiento del modelo a posteriori porque, aunque lo has construido con validación cruzada, y has escogido el que mejor se comporta con datos que "no ve", no sabes (quizá en tu caso sí, pero no

Re: [R-es] CV en R

Estimado Jesús Para Fernández Usted tiene unos datos que utilizó para entrenar, luego corrió dos modelos, y piensa que puede realizar otro modelo sin entrenamiento. Yo pienso que puede mejorar el modelo y realizar un entrenamiento a este nuevo modelo aunque sean los mismos datos, desconozco

Re: [R-es] CV en R

No, no. Si construyes el modelo con todos los datos, explícame para qué te ha servido la validación cruzada... ¿Sólo para saber si funciona mejor SVM o RF con ese conjunto de datos? Eso es insuficiente. Cuando construyes un modelo, lo haces entrenando con datos que el modelo NO VE, ahí está la

Re: [R-es] errores con openair (Javier Valdes Cantallopts (DGA))

Javier, parece que openair requiere un formato especifico de fecha y efectivamente "strptime". En la documentacion, funcion import: date.format The format of the date. This is given in ‘R’ format according to strptime. For example, a date format such as 1/11/2000 12:00 (day/month/year

Re: [R] Upper bands and lower bands

Sorry, For me a vector is a matrix with mx1 dimmensions. But it is true that it is not the way I correctly must talk in R. Can you guide me in what I´m trying to do? I´m trying to find what I want in Excel using something like a parabolic function but dind´t get yet (then to try to replicate in

[R] Question on interpreting glmer() results

Hello, I originally posted this on the stats stack exchange site, but given its focus on R software, it was removed -- so I figured I'd post here. I'm having trouble interpreting a change in effect direction and significance when I add an interaction term to my glmer() model. *Part 1* I ran an