The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
cpsrm(),
cpsrm_run())cpsrm(), a friendly entry point for the
Co-Partner SRM (CP-SRM) that takes raw long-format data plus
actor_id/group_id and builds the required
actor/partner dummy matrices internally via
create_cp_dummies(), then fits the model with
cpsrm_run().cpsrm_run(), the full-control fitting function
underneath cpsrm(), together with
create_cp_dummies(), print.cpsrm(), and
summary.cpsrm(). cpsrm_run() fits the CP-SRM –
in which each observation involves one actor and all other members of a
group acting simultaneously as partners (e.g. three-person teams) – via
REML, using either a Woodbury-matrix-identity (“block”) or
full-covariance-loop (“loop”) formulation, and supports variable group
sizes.print.cpsrm()/summary.cpsrm() report
variance-of-total percentages under BOTH the RAW convention (each
component’s own share of total variance; the correct basis for comparing
Partner’s effect size to Actor’s) and the COMBINED convention (Partner’s
contribution scaled by the number of partners summed into each row; the
correct basis for a full total-variance decomposition), whenever
weight_partners = FALSE and the number of partners per row
is constant.?cpsrm_run documents (a) the RAW vs. COMBINED
distinction and why conflating them can make partner effects look
several times more or less important than actor effects than they really
are, and (b) the boundary-corrected likelihood-ratio test procedure for
testing whether a variance component is zero (one-sided halved
chi-square for a single component; the three-part 1/4-1/2-1/4 mixture
for a joint two-component test), since variance components are
boundary-constrained and the naive two-sided/unhalved chi-square test is
not valid here.cpsrm()/cpsrm_run()/create_cp_dummies()
naming: createDummies() ->
create_dummies(), srmRun() ->
srm_run(), srmVarPct() ->
srm_var_pct(), srmPseudoRSq() ->
srm_pseudo_rsq()..Deprecated()
warning pointing to the new name and new (also renamed) arguments. No
user-facing breaking changes.createDummies().srmRun(), srmVarPct(),
srmPseudoRSq().nlme covariance class pdSRM
implementing the SRM variance-covariance constraints (equal actor
variances, equal partner variances, single actor-partner
covariance).sampleDyadData with simulated
round-robin data from two time points.groupId column in
srmRun() that used the variable name string rather than the
column contents when creating the pdSRM_group_id grouping
variable.class(x) == "try-error" with
inherits(x, "error") in pdMatrix.pdSRM() per
CRAN policy.stats:: namespace prefixes to calls to
formula(), na.omit(), and coef()
throughout.\donttest{} to comply
with CRAN example time limits.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.