TM 5-6675-323-14
Keystrokes
Display
L.R.
-118,290.6295
The y-intercept of the line.
61.1612
Slope of the line.
( 6 ) L i n e a r e s t i m a t i o n . W i t h d a t a a c c u m u l a t e d i n r e g i s t e r s R . 0 t h r o u g h R .5 a
predicted value for y (denoted y) can be calculated by keying in a new value for x
and
pressing
. A predicted value for x (denoted x) can be calculated by
keying in a new value for y and pressing
.
With data intact
from previous
e x a m p l e i n r e g i s t e r s R . 0 t h r o u g h R .5
Example:
t o predict demand for motor fuel
for the years
1980 and 2000, key in new x values
and
press
. T o determine
the year that
the demand for motor fuel is expected
to pass 3,500 million barrels,
key in 3,500
(new value for y) and press
Keystroke
Display
2,808.6264
Predicted demand in millions
of barrels for the year 1980.
4,031.8512
Predicted demand in millions
of barrels for the year 2000.
1,991.3041
T h e demand is expected to pass
3,500 million barrels during
1992.
(7) Correlation coefficient.
Both linear regression and linear estimation
p r e s u m e that the relationship between x and y data values can be approximated, to
some degree, by a linear function (a straight line).
(correlation coefficient)
can be used to determine how closely the data "fits" a straight line. The correla-
tion coefficient can range from r = + 1 to r = -1. A t r = + 1 , d a t a f a l l s e x a c t l y
onto a straight line with positive slope. While at r = -1, data falls exactly onto
a s t r a i g h t l i n e w i t h n e g a t i v e s l o p e . A t r = 0, data cannot be approximated by a
straight line.
To calculate the correlation coefficient for previous example press:
Example:
Keystrokes
Display
T h e data very closely
0.9931
approximates a straight
line.
only in a controlled environment.
13-34
