KALDFF Call
computes the one-step forecast of state
vectors in an SSM by using the diffuse Kalman filter.
The call estimates the conditional expectation of
, and also estimates the initial random
vector,
, and its covariance matrix.
- CALL KALDFF( pred, vpred, initial, s2, data, lead,
int, coef, var,
- intd, coefd <, n0, at, mt,
qt>);
The inputs to the KALDFF subroutine are as follows:
- data
- is a
matrix containing
data
.
- lead
- is the number of steps to forecast
after the end of the data set.
- int
- is an
matrix for a time-invariant
fixed matrix, or a
matrix containing fixed matrices for the time-variant
model in the transition equation and the measurement equation -
that is,
.
- coef
- is an
matrix for a time-invariant
coefficient, or a
matrix containing coefficients at each time in the
transition equation and the measurement equation -
that is,
.
- var
- is an
matrix for a
time-invariant variance matrix for the error in the transition
equation and the error in the measurement equation, or a
matrix
containing covariance matrices for the error in the transition
equation and the error in the measurement equation - that
is,
.
- intd
- is an
vector containing
the intercept term in the equation for the initial
state vector
and the mean effect
-
that is,
.
- coefd
- is an
matrix containing
coefficients for the initial state
in the equation
for the initial state vector
and the mean effect
- that is,
.

- is an optional scalar including an initial denominator.
If
, the denominator for
is
plus the number
of elements
of
.
If
or
is not specified, the
denominator for
is
.
With
, the initial values,
, and
, are assumed to be known and, hence,
,
,
and
are used for input containing the initial values.
If the value of
is negative or
is not specified,
the initial values for
,
, and
are computed.
The value of
is updated as
after the KALDFF call.

- is an optional
matrix.
If
,
contains
.
However, only the first matrix
is used as input.
When you specify the KALDFF call,
returns
.
If
is negative or the matrix
contains
missing values,
is automatically computed.

- is an optional
matrix.
If
,
contains
.
However, only the first matrix
is used as input.
If
is negative or the matrix
contains missing values,
is used for output,
and it contains
.
Note that the matrix
can be used as an input matrix
if either of the off-diagonal elements is not missing.
The missing element
is replaced
by the nonmissing element
.

- is an optional
matrix.
If
,
contains
.
However, only the first matrix
is used as input.
If
is negative or the matrix
contains missing values,
is used for
output and contains
.
The matrix
can also be used as an input
matrix if either of the off-diagonal elements is
not missing since the missing element
is replaced by the nonmissing element
.
The KALDFF call returns the following values:
- pred
- is a
matrix containing
estimated predicted state vectors
.
- vpred
- is a
matrix
containing estimated mean square errors of predicted state
vectors
.
- initial
- is an
matrix containing an
estimate and its variance for initial state
,
that is,
.

- is a scalar containing the estimated variance
.
The KALDFF call computes the one-step forecast of state
vectors in an SSM by using the diffuse Kalman filter.
The SSM for the diffuse Kalman filter is written

where

is an

state vector,

is an

observed vector, and
![[ \eta_t \ \epsilon_t ] & \sim & n (0, \sigma^2 [ {v}_t & {g}_t \ {g}^'_t & {{r}}_t ] ) \ \delta & \sim & n(\mu, \sigma^2\sigma)](images/langref_langrefeq606.gif)
It is assumed that the noise vector

is
independent and

is independent of the vector

.
The matrices,

,

,

,

,

,

,

,

,

,

, and

, are assumed to be known.
The KALDFF call estimates the conditional expectation
of the state vector

given the observations.
The KALDFF subroutine also produces the estimates of the
initial random vector

and its covariance matrix.
For

-step forecasting where

, the
estimated conditional expectation at time

is computed with observations given up to time

.
The estimated

-step forecast and its estimated MSE
are denoted

and

(for

).

and

are last-column-deleted submatrices of

and

, respectively.
The algorithm for one-step prediction is given as follows:

where

is the number of elements of

plus

.
Unless initial values are given and

, initial values are set as follows:

For

-step forecasting where

,

Note that if there is a missing observation, the
KALDFF call computes the one-step forecast for the
observation following the missing observation as
the two-step forecast from the previous observation.
An example that uses the KALDFF call is in the documentation for the
KALDFS call.