Page 18-7
•
Obtain single-variable information using:
‚Ù
@@@OK@@@
. Use Sample for
the Type of data set, and select all options as results. The results for this
example were:
Mean: 51.0406, Std Dev: .29.5893…, Variance: 875.529…
Total: 10208.12, Maximum: 99.35, Minimum: 0.13
This information indicates that our data ranges from values close to zero to
values close to 100. Working with whole numbers, we can select the range
of variation of the data as (0,100). To produce a frequency distribution we
will use the interval (10,90) dividing it into 8 bins of width 10 each.
•
Select the program
2. Frequencies..
by using
‚Ù˜
@@@OK@@@
. The
data is already loaded in
Σ
DAT, and the option Col should hold the value
1 since we have only one column in
Σ
DAT.
•
Change X-Min to 10, Bin Count to 8, and Bin Width to 10, then press
@@@OK@@@
.
Using the RPN mode, the results are shown in the stack as a column vector in
stack level 2, and a row vector of two components in stack level 1. The
vector in stack level 1 is the number of outliers outside of the interval where
the frequency count was performed. For this case, I get the values [ 25. 22.]
indicating that there are, in my
Σ
DAT vector, 25 values smaller than 10 and
22 larger than 90.
•
Press
ƒ
to drop the vector of outliers from the stack. The remaining
result is the frequency count of data. This can be translated into a table
as shown below.
This table was prepared from the information we provided to generate the
frequency distribution, although the only column returned by the calculator is
the Frequency column (f
i
). The class numbers, and class boundaries are easy
to calculate for uniform-size classes (or bins), and the class mark is just the
average of the class boundaries for each class. Finally, the cumulative
frequency is obtained by adding to each value in the last column, except the
first, the frequency in the next row, and replacing the result in the last column