stat_cell {etable}

Diverse statistics cell function

Description

Calculating values of several descriptive statistics.

Usage

       stat_cell(x, y, z, w, cell_ids, row_ids, col_ids, vnames, vars, n_min,
       digits = 3, digits2=1)

Arguments

x

The x variable

y

NOT USED

z

NOT USED

w

Weights for x variable.

cell_ids

Index vector for selecting values in cell.

row_ids

NOT USED

col_ids

NOT USED

vnames

NOT USED

vars

A vector of character strings with names of variables in data.frame for x, y and z. Use names of x or y as keywords, to choose a certain statistic.

n_min

Minimum n in the cell for useful calculation. Cells with n<n_min deliver no output.

digits

Integer indicating the number of significant digits.

digits2

Integer indicating the number of decimal places for percentages.

Details

Keywords are:

• N: number in this cell

• MIN: minimum

• MAX: maximum

• SUM: sum

• MEAN: mean

• SD: standard deviation

• MSD: mean, standard deviation

• MCI: mean, 95% CI

• VAR: variance

• MEDIAN: median

• MD: mean deviation from the mean (*1.253)

• MAD: median absolute deviation (*1.4826)

• IQR: interquartile range

• MQQ: median (Q1/Q3)

• PROP: proportion

• POP: proportion of level 2 (only binar)

• PCI: proportion of level 2, 95% CI

• RANGE: range

• CV: coefficient of variation

• MODE: mode

• MISS: number of missing values

• PNM: proportion of non missing values

• COMB: POP for binar and MQQ for continues

• SKEW: skewness

• KURT: excess kurtosis

• GEO: geometric mean

• HARM: harmonic mean

• TM1: truncated mean 1%

• TM5: truncated mean 5%

• TM10: truncated mean 10%

• TM25: truncated mean 25%

• WM1: winsorized mean 1%

• WM5: winsorized mean 5%

• WM10: winsorized mean 10%

• WM25: winsorized mean 25%

• M1SD: mean-SD, mean+SD

• M2SD: mean-2SD, mean+2SD

• M3SD: mean-3SD, mean+3SD

• MM1SD: mean, mean-SD, mean+SD

• MM2SD: mean, mean-2SD, mean+2SD

• MM3SD: mean, mean-3SD, mean+3SD

• NORM50: mean-0.675SD, mean+0.675SD

• NORM90: mean-1.645SD, mean+1.645SD

• NORM95: mean-1.96SD, mean+1.96SD

• NORM99: mean-2.576SD, mean+2.576SD

• P1: 1th quantile

• P2.5: 2.5th quantile

• P5: 5th quantile

• P10: 10th quantile

• P20: 20th quantile

• P25: 25th quantile

• P30: 30th quantile

• P40: 40th quantile

• P50: 50th quantile

• P60: 60th quantile

• P70: 70th quantile

• P75: 75th quantile

• P80: 80th quantile

• P90: 90th quantile

• P95: 95th quantile

• P97.5: 97.5th quantile

• P99: 99th quantile

Author(s)

Andreas Schulz <[e-mail]>

Examples

sex     <- factor(rbinom(1000, 1, 0.4),  labels=c('Men', 'Women'))
height  <- rnorm(1000, mean=1.66, sd=0.1)
height[which(sex=='Men')]<-height[which(sex=='Men')]+0.1
weight  <- rnorm(1000, mean=70, sd=5)
decades <- rbinom(1000, 3, 0.5)
decades <- factor(decades, labels=c('[35,45)','[45,55)','[55,65)','[65,75)'))
d<-data.frame(sex, decades, height, weight)


tabular.ade(x_vars=c('height', 'weight'), xname=c('Height [m]','Weight [kg]'),
   y_vars=c('N', 'MEAN', 'SD', 'SKEW', 'KURT'),
   rows=c('sex', 'ALL', 'decades', 'ALL'), rnames=c('Gender', 'Age decades'),
   data=d, FUN=stat_cell)