Page 18-14
•
To access the
summary stats…
option, use:
‚Ù˜˜˜
@@@OK@@@
•
Select the column numbers corresponding to the x- and y-data, i.e., X-Col:
1, and Y-Col: 2.
•
Using the
@
CHK@
key select all the options for outputs, i.e., _
Σ
X, _
Σ
Y, etc.
•
Press
@@@OK@@@
to obtain the following results:
Σ
X: 24.2,
Σ
Y: 11.72,
Σ
X2: 148.54,
Σ
Y2: 26.6246,
Σ
XY: 12.602, N
Σ
:8
Note:
There are two other applications under the STAT menu, namely,
5.
Hypth. tests..
and
6. Conf. Interval..
These two applications will be discussed
later in the chapter.
Calculation of percentiles
Percentiles are measures that divide a data set into 100 parts. The basic
procedure to calculate the 100
⋅
p-th Percentile (0 < p < 1) in a sample of size
n is as follows:
1. Order the n observations from smallest to largest.
2. Determine the product n
⋅
p
A. If n
⋅
p is not an integer, round it up to the next integer and find the
corresponding ordered value.
B. If n
⋅
p is an integer, say k, calculate the mean of the k-th and (k-1) th
ordered observations.
Note
: Integer rounding rule, for a non-integer x.yz…, if y
≥
5, round up to
x+1; if y < 5, round up to x.
This algorithm can be implemented in the following program typed in RPN
mode (See Chapter 21 for programming information):
«
SORT DUP SIZE
p X n
«
n p *
k
«
IF k CEIL k FLOOR - NOT THEN X
k GET X k 1 + GET + 2 / ELSE k 0 RND X SWAP GET END
»
»
»
