Description Usage Arguments Details Value Examples
Convert (by numericising, centring or standardising) the variables of a data.frame or elements of a list into a list of integer or numeric values ready for analysis by JAGS, OpenBUGS or WinBUGS.
1 2 |
data |
the data frame or list of data on which the conversion is based |
numericise |
a logical scalar or a character vector of the variables to numericise |
centre |
a logical scalar or a character vector of the variables to centre |
standardise |
a logical scalar or a character vector of the variables to standardise |
dat |
the data frame or data list to convert. If dat is NULL the dataset passed as the data argument is converted |
Numericising, centring and standardising represent increasing conversion of a variable. A variable is first converted into a numeric value (numericising). Next the mean is subtracted from all the values (centring). Finally all the values are divided by the standard deviation (standardising).
Numericising has no effect on numeric and integer values. Factors are numericized by converting to an integer. Date and POSIXt variables are numericised by converting to integer values and subtracting the integer value of 1999-12-31 or 1999-12-31 23:59:59 UTC, respectively.
Centring has no affect on factors while integer, Date and POSIXt values are subtracted from the rounded mean. Similarly, standardising has no affect on factors while centred integer, Date and POSIXt values are divided by their respective standard deviation.
The converted data frame
1 2 3 4 5 6 7 8 | data <- data.frame(numeric = 1:10 + 0.1, integer = 1:10,
factor = factor(1:10), date = as.Date("2000-01-01") + 1:10,
posixt = ISOdate(2000,1,1) + 1:10)
convert_data(data)
convert_data(data, centre = TRUE)
convert_data(data, standardise = TRUE)
convert_data(data, numericise = FALSE, standardise = "date")
convert_data(data, numericise = FALSE, centre = "date", standardise = "date")
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.