search for: jimmycloud

Dear All, Suppose I have a categorical variable a=as.factor(sample(1:3,10,replace=T)) plot(a) and hist(as.numeric(a),freq=F) would give the histogram of it. But I do not know how to add the counts or percentage information for plot.factor(). hist() can do it but as a numeric variable, the x-axis is not 3 categories in this case. Thank you for any suggestion. Best wishes, Jie

Dear All, I have a question, suppose X is a dataframe, with column names as "x1", "x2", "x3", ..... And I would like to use the i-th column by X[,'xi']. But it seems the single quote and double quote are different. So if I run X[, names(X)[i]], it has some error. Please use the below example code X = matrix(rnorm(50),ncol = 5) X = data.frame(X)

Hi All, I have millions of regression lines to fit. So I am looking for the most efficient approach in R. Details: I have a large desing matrix X. The dimension is n by p. Each time when fitting the model, select rows from this matrix X and form a new design matrix, called X_current. There is another binary matrix M, with dim m by n, and each row is a 1*n vector. It helps to determin X_current.

Dear All, I am running a r code on 32bit win, involving absolutely small numbers. Although I tried sth like the ratio of numers like 10^(-100) and did not have issue to get the correct answer, but still a little concerned about it. Could anyone give some suggestion or have any experience? Best wishes, Jie [[alternative HTML version deleted]]

Dear All, This might be a tiny question but I do not know the reason. On my desktop, sometimes when I use the mouse or ctrl+A select a piece of R script, and use ctrl+R to run it, there is no action. If I redo it, it may work. On my laptop, it is fine. Both are running windows 7. Best wishes, Jie [[alternative HTML version deleted]]

Hi All, I have a txt file to read into R. The size of it is about 500MB. This txt file is produced by calling write.table(M, file = "xxx.txt"), where M is a large matrix After running MM = read.table("xxx.txt"), the R gui keeps a cpu core/thread fully occupied forever. 64 bit R with 16GB RAM on Win7 64, i5 cpu should be capable. So if anyone knows the reason, that will be

Dear All, I have a long sequence and want to find the quantile, or sort it first. It seems sort() or quantile() reaches the memory limit. Is there a way to allocate more memoy on SSD for R when startup, so that R can use both RAM and hard drive space? Thank you. Best wishes, Jie

Dear All, I would like to store quote as part of an vector. For instance, I would like to get an character object as x = " 12"ab"34 " or y = c("1", "2", """, "a", "b", """, "3", "4") Is that possible? Thank you. Best wishes, Jie [[alternative HTML version deleted]]

Dear All, I performed thousands of testings and obtained p-values. And then I did two-sided uniform KS test of the p-values, the result claimed it is uniform. So does it mean that my model are wrong? Because I expect more small p-values near 0. This is a preliminary step before correcting the multiplicity. Attached is hist of p-values (does this list allow attachment?). The ks test: One-sample

Dear All, Suppose I have a program as below: Outside is a loop for simulation (with random generated data), inside there are several sapply()'s (10~100) over the data and something else, but these sapply's have to be sequential. And each sapply do not involve very intensive calculation (a few seconds only). So the outside loop takes minutes to finish one iteration. I guess the better way

Dear All, I am learning parallel in R and start with the package "snow". I did a test about running time and the parallel version is much slower than the regulat code. My laptop is X200s with dual core intel L9400 cpu. Should I make more clusters than 2? Or how to improve the performance? # install.packages("snow") library(snow) cl <- makeCluster(2) t1 <- proc.time() a

Dear All, I have three classes of questions about generating random numbers with different packages (windows xp 32bit R). . 1. Suppose I would like to use package *foreach*, can I use current Sys.time as a seed? Although I can get the time up to1e-6 second precesion, the code below dose not work well on a local machine with two cores. ################# library(foreach) library(snow)

I have a general question about coefficients estimation of the mixed model. I simulated a very basic model: Y|b=X*\beta+Z*b +\sigma^2* diag(ni); b follows N(0,\psi) #i.e. bivariate normal where b is the latent variable, Z and X are ni*2 design matrices, sigma is the error variance, Y are longitudinal data, i.e. there are ni

Dear All, I am curious if there is any why to use R to know the specification of the machine that runs on, i.e. read the cpu model, memory size, those hardware info, maybe even with thesoftware information. Thank you for your attention. Best wishes, Jie [[alternative HTML version deleted]]

Dear All, I am evaluating the value of loglikelihood and it ends up with the sum of tiny numbers. Below is an example: suppose I would like to calculate sum_i (log (sum_j x [i, j] )), the index of log (x) is in the range, say (-2000, 0). I am aware that exp(-744.5) will be expressed as 0 in 32 bit R and exp Is there a way to improve the result? R example: powd <- sample(-2000:0, 100,

Dear All, My question is simple but I need someone to help me out. Suppose I have a positive definite matrix A. The funtion chol() gives matrix L, such that A = L'L. The inverse of A, say A.inv, is also positive definite and can be factorized as A.inv = M'M. Then A = inverse of (A.inv) = inverse of (M'M) = (inverse of M) %*% (inverse of M)' = ((inverse of

Dear All, Suppose I have a vector X = (x_1, x_2, ...., x_n), X_sort = sort(X) = (x_(1), x_(2), ... , x(n) ), and I would like to know the original position of these ordered x_(i) in X, how can I do it? case 1: all values are unique x <- c( 3, 5, 4, 6) x.sort <- sort(x) # # I would like to obtain a vector (1, 3, 2, 4) which indicates that 3 in x is still the 1st element in x.sort, 5 is at

Dear All, have a general question about coefficients estimation of the mixed model. I simulated a very basic model: Y|b=X*\beta+Z*b +\sigma^2* diag(ni); b follows N(0,\psi) #i.e. bivariate normal where b is the latent variable, Z and X are ni*2 design matrices, sigma is the error variance, Y are longitudinal data, i.e. there are ni