Re: [R] Doubt_merging data

2018-04-09 Thread MacQueen, Don
Your email is hard to read because you sent html email. Please send plain text. You will need to say what you mean by "join". It's not a standard term with a universally agreed upon meaning within R. If you have 5 data frames, each with 5 rows, and it makes sense that after joining you should

Re: [R] convert numeric variables to factor

2018-04-09 Thread Saif Tauheed
Dear Sir, I have xlsx data set which I have imported to R studio. Now some of the variables are defined as numeric but I want define them as factor variable so that I run classification algorithm in R. Please to covert the variables. Thanks and Regards Abu Afzal PhD Eco JNU India

Re: [R] convert numeric variables to factor

2018-04-09 Thread David L Carlson
Try the help files: ?factor David L Carlson Department of Anthropology Texas A University College Station, TX 77843-4352 -Original Message- From: R-help On Behalf Of Saif Tauheed Sent: Monday, April 9, 2018 11:29 AM

[R] Specifying forbidden configurations in Morris One at a Time (OAT) sensitivity analysis

2018-04-09 Thread Reshma C
Hi! I am trying to implement the Morris One at a Time (OAT) sensitivity analysis technique in R using the package 'sensitivity'. The OAT and similar other techniques in the package requires data frame X (which stores the design or parameter combinations) to be populated by some function in the

Re: [R] convert numeric variables to factor

2018-04-09 Thread Rui Barradas
Hello, Though Bert's and David's answers are what you should do, note that some R functions that need factors will coerce their input variables when necessary. Have you tried to run the code you haven't posted without coercing to factor? It might run... Hope this helps, Rui Barradas On

Re: [R] convert numeric variables to factor

2018-04-09 Thread Bert Gunter
Just cast it! ?factor Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Mon, Apr 9, 2018 at 9:28 AM, Saif Tauheed wrote: > Dear Sir, > >

Re: [R] convert numeric variables to factor

2018-04-09 Thread Bert Gunter
see also ?cut if this is what you mean. -- Bert Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Mon, Apr 9, 2018 at 9:28 AM, Saif Tauheed

Re: [R] convert numeric variables to factor

2018-04-09 Thread William Dunlap via R-help
Or use cut(): > Num <- c(2.2, 2.4, 3.5, 5, 7) > cut(Num, breaks=c(0,2,4,6), labels=c("Low","Medium","High")) [1] Medium Medium Medium High Levels: Low Medium High Bill Dunlap TIBCO Software wdunlap tibco.com On Mon, Apr 9, 2018 at 10:00 AM, Bert Gunter wrote: >

Re: [R] convert numeric variables to factor

2018-04-09 Thread Samantha Smith
i am seriously meet up with you. if you're interested in getting to know me Then sent me about your details . William Im really serious! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see

[R] Fail to save an object using name as string

2018-04-09 Thread Marc Girondot via R-help
Dear list member, I think that I have detected a strange behavior of the save() command: > year <- "2000" > assign(paste0("Var_", year), list(A=10, B=20)) > get(paste0("Var_", year)) $A [1] 10 $B [1] 20 # At this point all is ok, I have created a list of name Var_2000 > save(paste0("Var_",

Re: [R] linear regression with repeated measures

2018-04-09 Thread Bert Gunter
Generally, statistics questions are off topic here, although they do sometimes intersect R programming issues, as perhaps here. Nevertheless, I believe your post would fit better on the r-sig-mixed-models list, where repeated measures and other mixed effects (/variance components) models are

Re: [R] convert numeric variables to factor

2018-04-09 Thread Samantha Smith
Cool  William i'm 100 sure you and me we can have a good time together. Just trust me okey.. Am not an escort ok so plz dont think me like that. I just need you to prove you are not minor and safe. My friends also here with me while am writing you. If you can come now you can ride us both.

[R] linear regression with repeated measures

2018-04-09 Thread Wolfgang Raffelsberger
Dear list, this seemed to me like a very trivial question, but finally I haven't found any similar postings with suitable solutions on the net ... Basically, instead of regressing two simple series of measures 'a' and 'b' (like b ~ a), I would like to use independent replicate measurements for

[R-es] Warning en modelo ZINB

2018-04-09 Thread miriam . alzate
Buenas tardes, Estoy estimando un modelo binomial negativo de ceros inflados (ZINB) utilizando el comando zeroinfl() del paquete pscl. Al ejecutarlo me da el siguiente aviso: Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred ¿Sabéis que significa y si puedo usar el modelo aún

Re: [R-es] Warning en modelo ZINB

2018-04-09 Thread Carlos J. Gil Bellosta
Hola, ¿qué tal? El "warning" que comentas aparece en glm.fit precisamente cuando un coeficiente diverge. El aviso puede ser malo o irrelevante, depende. Puede que haya sido emitido en algún paso intermedio del ajuste (por lo que no habría mayor problema). O que afecte al ajuste entero. Uno de

Re: [R-es] Warning en modelo ZINB

2018-04-09 Thread miriam . alzate
Muchas gracias por la respuesta. He mirado y los coeficientes no son altos pero sí tengo una gran cantidad de ceros en la variable dependiente (más del 90%). Sin embargo, al incluir otro tipo de variables independientes no me da ese aviso, dejando la misma variable dependiente. ¿Cómo podría

Re: [R-es] Warning en modelo ZINB

2018-04-09 Thread miriam . alzate
¿Quieres decir que para un nivel de una variable categorica todas las observaciones de la variable respuesta sean ceros? Gracias El Lun, 9 de Abril de 2018, 19:59, Carlos J. Gil Bellosta escribió: > ¿Podría ser que para algún nivel de alguna variable independiente > categórica solo hubiese

Re: [R-es] Warning en modelo ZINB

2018-04-09 Thread miriam . alzate
Hola de nuevo Carlos, he probado a quitar esa variable categórica y me sigue dando el aviso... El Lun, 9 de Abril de 2018, 20:17, Carlos J. Gil Bellosta escribió: > Si, creo que el motivo del warning puede ser ese. Es hipotético, pero > plausible. Sobre todo cuando tienes más de un 90% de ceros.

Re: [R-es] Warning en modelo ZINB

2018-04-09 Thread miriam . alzate
En ese caso, ¿tendría sentido el modelo? o ¿debería quitar esa variable categórica? Muchas gracias El Lun, 9 de Abril de 2018, 20:17, Carlos J. Gil Bellosta escribió: > Si, creo que el motivo del warning puede ser ese. Es hipotético, pero > plausible. Sobre todo cuando tienes más de un 90% de

Re: [R-es] Warning en modelo ZINB

2018-04-09 Thread Carlos J. Gil Bellosta
¿Podría ser que para algún nivel de alguna variable independiente categórica solo hubiese ceros? En ese caso, casi seguro, aparecería ese tipo de warning. El lun., 9 abr. 2018 a las 19:00, escribió: > Muchas gracias por la respuesta. He mirado y los coeficientes no

Re: [R-es] Warning en modelo ZINB

2018-04-09 Thread Carlos J. Gil Bellosta
Si, creo que el motivo del warning puede ser ese. Es hipotético, pero plausible. Sobre todo cuando tienes más de un 90% de ceros. El coeficiente de ese nivel para el modelo de la mixtura (ceros vs binomial negativa) sería infinito. Y de ahí el warning. El lun., 9 abr. 2018 a las 20:09,