About the format of the data files.
The data sets that supplement the volume are usually in the form of
a matrix or table. Each table will appear as a separate file and the
description will contain information about the number of rows and columns.
Missing values appear as NA
Some details:
- Stacked column format
Many of the data sets have too many columns to list them easily row
by row so a different format is used to store them in files.
The text files containing the data will have one number per line where
the values are ordered column by column (see the example below).
As a check, the begining the first row will be listed in the description
so that you can check that the table is formated correctly.
(This is the convention for storing the table is the same S and FORTRAN.)
An example of file format:
The
table ( i.e 3X2 matrix) data set
1 2 3 NA
5 6 7 8
9 10 11 12
would be strung out in file on 12 lines as:
1
5
9
2
6
10
3
7
11
NA
8
12
- S data set objects.
To read the S data
sets using S-PLUS you should get the dumped
S-object. (To make it easier this object may actually contain several of
the individual datasets bundled together as a
list.) If you put this in the file say ozone.dat then in S-PLUS
source("ozone.dat") will read in the dataset.
