How can i estimate the median in R using the median fuction?

Question

i am trying to calculate some statistical moments, using the moments function, for a group of variables. In order to do this i try to program a function

When I try to use my function with my dataframe that does not have NA or missing, I confirmed that the data is clean, I have the next error:

library(moments)

summary.stat0 = function(data){ 
  SUM = matrix(NA,ncol=ncol(data),nrow=7)
  colnames(SUM) = colnames(data) 
  rownames(SUM) = c("Mean","Median","Maximum","Minimum","Std. Dev.","Skewness","Kurtosis")
  for (i in 1:ncol(data)) {
    SUM[1,i] = mean(data[,i])
    SUM[2,i] = median(data[,i])
    SUM[3,i] = max(data[,i])
    SUM[4,i] = min(data[,i])
    SUM[5,i] = sd(data[,i])
    SUM[6,i] = skewness(data[,i])
    SUM[7,i] = kurtosis(data[,i])
  }
  SUM
}

Error in median.default(data[i]): need numeric data In addition: Warning message: In mean.default(data[, i]):

If someone can help me with this function i will be grateful

Answer 1

You can compute all needed statistics in one apply loop.

library(moments)

summary.stat <- function(data, na.rm = TRUE){ 
  SUM <- apply(data, 2, function(x){
    c(Mean = mean(x, na.rm = na.rm),
      Median = median(x, na.rm = na.rm),
      Max = max(x, na.rm = na.rm),
      Min = min(x, na.rm = na.rm),
      `Std. Dev.` = sd(x, na.rm = na.rm),
      Skewness = skewness(x, na.rm = na.rm),
      Kurtosis = kurtosis(x, na.rm = na.rm)
    )})
  colnames(SUM) <- colnames(data) 
  SUM
}

set.seed(1234)
df1 <- as.data.frame(replicate(3, rnorm(10)))

summary.stat(df1)
#                  V1         V2          V3
#Mean      -0.3831574 -0.1181707 -0.38794682
#Median    -0.5555419 -0.4941011 -0.46561688
#Max        1.0844412  2.4158352  0.57475572
#Min       -2.3456977 -0.9983864 -1.44820491
#Std. Dev.  0.9957875  1.0673376  0.66600127
#Skewness  -0.4243347  1.5011660  0.02058876
#Kurtosis   2.6686992  4.1925245  1.87910667