E-25
Calculating Estimated Values
Based on the regression formula obtained by paired-variable statistical
calculation, the estimated value of
y can be calculated for a given x-value.
The corresponding
x-value (two values, x
1
and x
2
, in the case of quadratic
regression) also can be calculated for a value of
y in the regression
formula.
To determine the estimate value for y when x = 160 in the
regression formula produced by logarithmic regression of the data
in
3
. Specify Fix 3 for the result. (Perform the following operation
after completing the operations in
3
.)
160 (STAT)(Reg)(L)
Result:
8106.898
Important: Regression coefficient, correlation coefficient, and estimated
value calculations can take considerable time when there are a large number
of data items.
Performing Normal Distribution Calculations
While single-variable statistical calculation is selected, you can perform
normal distribution calculation using the functions shown below from
the menu that appears when you perform the following key operation:
(STAT)(Distr).
P, Q, R: These functions take the argument
t and determine a probability of
standard normal distribution as illustrated below.
t: This function is preceded by the argument X, and determines the
normalized variate .
For the single variable data {x
n
; freq
n
} = {0;1, 1;2, 2;1, 3;2, 4;2, 5;2,
6;3, 7;4, 9;2, 10;1}, to determine the normalized variate (t) when x
= 3, and P(t) at that point up to three decimal places (Fix 3).
(SETUP)A(STAT)(ON)
(SETUP)(Fix)(STAT)(1-VAR)
0
1 2 3 4 5 6 7 9
10AC1212223
4
2 1
3 (STAT)(Distr)(
t)
44
P
(t)Q
(t)R
(t)
0 t 0 t 0 t
P
(t)Q
(t)R
(t)
0 t 0 t 0 t
55
STAT
FIX
STAT
FIX
STAT
FIX
STAT
FIX