(Regression)
to
•
𝑥
,
𝑥
1
,
𝑥
2
and
𝑦
are commands of the type that take an argument
immediately before them.
Ex 2:
To input the single-variable data
x
= {1, 2, 2, 3, 3, 3, 4, 4, 5}, using the
Freq column to specify the number of repeats for each items {
x
n
; freq
n
} =
{1;1, 2;2, 3;3, 4;2, 5;1}, and calculate the mean.
(SETUP)
*1
or
*2
(Statistics)
(On)
*1: fx-82AR X *2: fx-95AR X
(Select Type)
(1-Variable)
1
2
3
4
5
1
2
3
2
(Variable)
(
𝑥
−
)
3
Ex 3:
To calculate the logarithmic regression correlation coefficients for the
following paired-variable data and determine the regression formula: (
x
,
y
) =
(20, 3150), (110, 7310), (200, 8800), (290, 9310). Specify Fix 3 (three
decimal places) for results.
(SETUP)
*1
or
*2
(Statistics)
(Off)
*1: fx-82AR X *2: fx-95AR X
(SETUP)
(Number Format)
(Fix)
(Select Type)
(y=a+b
･
ln(x))
20
110
200
290
3150
7310
8800
9310
(Regression)
(r)
0.998
(Regression)
(a)
-3857.984
(Regression)
(b)
2357.532
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.
Ex 4:
To determine the estimate value for
y
when
x
= 160 in the regression
formula produced by logarithmic regression of the data in Ex 3. Specify Fix
3 for the result. (Perform the following operation after completing the
operations in Ex 3.)
160
(Regression)
(
𝑦
)
8106.898
Important:
Regression coefficient, correlation coefficient, and estimated
value calculations can take considerable time when there are a large
number of data items.
Equation Calculations (fx-95AR X only)
Perform the steps below to solve an equation in the Equation/Func Mode.
1. Press
, select the Equation/Func Mode icon, and then press
.
2. Select the type of calculation you want to perform.
To select this calculation
type:
Do this:
17