Texas Instruments TI-73 Benutzerhandbuch

Seite von 364
Chapter 7: Statistical Analyses     
137
7307ENG.DOC   CH 7, Statistical Analyses, English   Julie Hewlett   Revised: 07/29/98 12:07 PM   Printed: 05/19/99
9:00 AM   Page 137 of 22
Variables
Definition
VARS Menu
r
Correlation coefficient
EQ
r
2
 or R
2
Coefficient of determination
EQ
RegEQ
Regression equation
EQ
x1,y1,x2,y2,
x3,y3
Summary points
PTS
a, b, c
Regressionà fit coefficients
EQ
 
n (number of data points)
n
=number of 
x
 data points in a 
1-Var Stats
 analysis or the
number of 
x
 and 
y
 data points in a 
2-Var Stats
 analysis. Since
both variable lists always have the same number of list
elements in 
2-Var Stats
n
 for 
x
 is always equal to 
n
 for 
y
.
Therefore, 
n
 applies to both the 
x
 and 
y
 analyses.
 
freq (Frequency Lists)
If freq is specified, 
n
 is equal to the sum of the elements in that
list. For example, if the freq is {2,2,3,1,2}, 
n
={2+2+3+1+2}=10.
 
Q
 
1
 
, Q
 
3
 
, and Med
Q
1
Q
3
, and 
Med
 are undefined if the freq contains non-integer
values. They also are not calculated if the freq contains a value
larger than 99.
 
RegEQ
The calculator stores the most recently generated regression
equation (see - v 
CALC
 menu items 3–7) to the variable,
RegEQ
. If, for example, you execute 
5: LinReg(ax+b)
, but you
don’t initially store 
RegEQ
 to a 
Y
n
 variable, you can later insert
RegEQ
 into the Y= editor. The calculator graphs the regression
equation when it is selected.
If the frequency for an element or data pair is 0, the element or
data pair is ignored in the calculation.