g02jgc combines output from multiple calls to
g02jfc or
g02jcc.
Let
and
denote two sets of data, each with
variables and
and
observations respectively. Let
and
denote two sets of communication arrays constructed by
g02jfc or
g02jcc from datasets
and
respectively. Then, given
and
,
g02jgc constructs a set of communication arrays,
, as if a dataset
, with
variables and
observations were supplied to
g02jfc or
g02jcc, with
constructed as
Splitting, and then recombining, the data in this manner allows for datasets with an arbitrarily large number of observations (
) to be analysed and the preprocessing routines,
g02jfc or
g02jcc, to be run in parallel.
It should be noted that, while the information in
, should be consistent with the information in the communication arrays obtained by supplying
to
g02jfc or
g02jcc, the ordering of that information may change. In practice, this means that whilst an analysis run using a set of communication arrays constructed using
g02jgc should give similar results to an analysis run using a set of communication arrays constructed directly from
g02jfc or
g02jcc they will not necessarily be identical. In addition, the order of the parameter estimates,
and
may differ.
None.
Not applicable.
Background information to multithreading can be found in the
Multithreading documentation.
The data preprocessing routines,
g02jcc and
g02jfc return the quantities,
fnlsv,
nff,
rnlsv,
nrf and
nvpr. These values are data dependent and, therefore, the values obtained when preprocessing the datasets,
and
are likely to be different to those obtained if you had preprocessed the combined dataset
. In order to obtain these values for
you can call
g02zlc after the call to
g02jgc. Valid values for
optstr are:
- FNLSV, in which case g02zlc will return fnlsv for the combined dataset .
- NFF, in which case g02zlc will return nff for the combined dataset .
- RNLSV, in which case g02zlc will return rnlsv for the combined dataset .
- NRF, in which case g02zlc will return nrf for the combined dataset .
- NVPR, in which case g02zlc will return nvpr for the combined dataset .
This example fits a random effects model to a simulated dataset. The dataset is split into four blocks, with
,
and
observations respectively. Each block is read in, processed by
g02jfc and then combined using
g02jgc. Once all four blocks have been processed the model is fitted using
g02jhc and the results are printed.