normalize_df_per_dim: Useful functions on data frames

In YAPSA: Yet Another Package for Signature Analysis

Description

normalize_df_per_dim: Normalization is carried out by dividing by rowSums or colSums; for rows with rowSums=0 or columns with colSums=0, the normalization is left out.

average_over_present: If averaging over columns, zero rows (i.e. those with rowSums=0) are left out, if averaging over rows, zero columns (i.e. those with colSums=0) are left out.

sd_over_present: If computing the standard deviation over columns, zero rows (i.e. those with rowSums=0) are left out, if computing the standard deviation over rows, zero columns (i.e. those with colSums=0) are left out.

stderrmean_over_present: If computing the standard error of the mean over columns, zero rows (i.e. those with rowSums=0) are left out, if computing the standard error of the mean over rows, zero columns (i.e. those with colSums=0) are left out. Uses the function stderrmean

Usage

normalize_df_per_dim(in_df, in_dimension)

average_over_present(in_df, in_dimension)

sd_over_present(in_df, in_dimension)

stderrmean_over_present(in_df, in_dimension)

Arguments

in_df	Data frame to be normalized
in_dimension	Dimension along which the operation will be carried out

Value

The normalized numerical data frame (normalize_df_per_dim)

A vector of the means (average_over_present)

A vector of the standard deviations (sd_over_present)

A vector of the standard errors of the mean (stderrmean_over_present)

See Also

stderrmean

Examples

test_df <- data.frame(matrix(c(1,2,3,0,5,2,3,4,0,6,0,0,0,0,0,4,5,6,0,7),
                             ncol=4))
## 1. Normalize over rows:
normalize_df_per_dim(test_df,1)
## 2. Normalize over columns:
normalize_df_per_dim(test_df,2)

test_df <- data.frame(matrix(c(1,2,3,0,5,2,3,4,0,6,0,0,0,0,0,4,5,6,0,7),
                             ncol=4))
## 1. Average over non-zero rows:
average_over_present(test_df,1)
## 2. Average over non-zero columns:
average_over_present(test_df,2)

test_df <- data.frame(matrix(c(1,2,3,0,5,2,3,4,0,6,0,0,0,0,0,4,5,6,0,7),
                             ncol=4))
## 1. Compute standard deviation over non-zero rows:
sd_over_present(test_df,1)
## 2. Compute standard deviation over non-zero columns:
sd_over_present(test_df,2)

test_df <- data.frame(matrix(c(1,2,3,0,5,2,3,4,0,6,0,0,0,0,0,4,5,6,0,7),
                             ncol=4))
## 1. Compute standard deviation over non-zero rows:
stderrmean_over_present(test_df,1)
## 2. Compute standard deviation over non-zero columns:
stderrmean_over_present(test_df,2)

YAPSA documentation built on Nov. 8, 2020, 4:59 p.m.