Statistics 12-27
8312STAT.DOC TI-83 international English Bob Fedorisko Revised: 02/19/01 12:42 PM Printed: 02/19/01 1:37
PM Page 27 of 38
PwrReg
(power regression) fits the model equation y=ax
b
to
the data using a least-squares fit and transformed values
ln(x) and ln(y). It displays values for
a
and
b
; when
DiagnosticOn
is set, it also displays values for
r
2
and
r
.
PwrReg
[
Xlistname
,
Ylistname
,
freqlist
,
regequ
]
Logistic
fits the model equation y=c / (1+a
ä
e
L
bx
) to the data
using an iterative least-squares fit. It displays values for
a
,
b
,
and
c
.
Logistic
[
Xlistname
,
Ylistname
,
freqlist
,
regequ
]
SinReg
(sinusoidal regression) fits the model equation
y=a sin(bx+c)+d to the data using an iterative least-squares
fit. It displays values for
a
,
b
,
c
, and
d
. At least four data
points are required. At least two data points per cycle are
required in order to avoid aliased frequency estimates.
SinReg
[
iterations
,
Xlistname
,
Ylistname
,
period
,
regequ
]
iterations
is the maximum number of times the algorithm
will iterate to find a solution. The value for
iterations
can
be an integer
‚
1 and
16; if not specified, the default is 3.
The algorithm may find a solution before
iterations
is
reached. Typically, larger values for
iterations
result in
longer execution times and better accuracy for
SinReg
, and
vice versa.
A
period
guess is optional. If you do not specify
period
, the
difference between time values in
Xlistname
must be equal
and the time values must be ordered in ascending
sequential order. If you specify
period
, the algorithm may
find a solution more quickly, or it may find a solution when
it would not have found one if you had omitted a value for
period
. If you specify
period
, the differences between time
values in
Xlistname
can be unequal.
Note: The output of
SinReg
is always in radians, regardless of the
Radian/Degree
mode setting.
A
SinReg
example is shown on the next page.
PwrReg
(ax
b
)
Logistic
c / (1+a
ä
e
L
bx
)
SinReg
a sin(bx+c)+d
