The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
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[@]dogado.de.
OptStiefelGBB.R.Tensor Regression with Envelope Structure and Three Generic Envelope Estimation Approaches'' toTensor
Regression with Envelope Structure’’.res <- kroncov(En) to
res <- try(kroncov(En)) in function
TRR.fit, which is forgotten to be reverted in last
version.base::norm(x, 'F') function with
sqrt(sum(x^2)) for some matrix x.Data segment of the printout of the
tensor object, instead of just the first few elements, the
summarized results of all elements are printed.OptManiMulitBallGBB() to the output of
ballGBB1D().TRR_sim -> TRRsim; TPR_sim -> TPRsim; TensEnv_dim -> TRRdim; ballGBB1D_bic -> oneD_bic; TensPLS_cv2d3d -> TPRdim; OptimballGBB1D -> OptM1D; EnvMU -> simplsMUplot.Tenv:
xlab = "", ylab = "",
axes = TRUE, ask = TRUE, remove arguments
xticks, yticks. axes is a logical
value specifying whether the axes should be drawn. If
ask = TRUE, user is prompted before the second plot is
shown (if exists).ECD, simplsMU, manifold1D, manifoldFG, OptM1D, OptMFG:
Change optional arguments like maxiter, tol,
to three dots ...fun1D, get_ini1D, ballGBB1D, OptManiMulitBallGBB:
Collected into file “1Dfunction.R”. Help documentation are no longer
supported.MenvU_sim: An option wishart = FALSE is
provided to return the population matrices M and
U. Other argument like jitter is also provided
which adds an scaler matrix to M to ensures it
positive-definiteness.PMSE: Calculates the prediction and mean squared error
for both TRR and TPR models.kroncov: Add convergence criterion tol and
the maximal iteration maxiter.OptStiefelGBB: Hide out from the
output.manifoldFG: Set default value for
Gamma_init.ttt.R: Change arguments X,Y to
x,y.oneD_bic: Add the estimated envelope basis
Gamma to the output.TRRdim: Output the mean squared error using the
selected envelope basis.TRRdim, oneD_bic: Argument multiD is
changed to C to comply with the paper Zhang X, Mai Q
(2018). “Model-Free Envelope Dimension Selection.” Electronic Journal of
Statistics.OptMFG: New FG optimization function encapsulating the
core function OptStiefelGBB.show: overloads show in package
rTensor. With the overloaded show, only the
first 6 elements of tensor is printed out.subspace(), the formula of subspace
distance should be ||P_{A} - P_{B}||_F/√{2d}.x was not binary.TRR.fit() and TPR.fit(). They can be passed
to argument x.Tenv class object, we remove S3 methods
vcov.Tenv(). Use function std_err() if the
standard error for the tensor coefficient from TRR.fit() is
desired.EEG. Refer to R help documentation
for more detailspredict.Tenv(): if the argument
newdata is missing, the fitted values from the fitted model
is returned.Xn and Yn in all
functions to x and y in accordance with other
popular functions, e.g., lm(), glm(),
etc.plot.Tenv(): Change the name of argument
thrd to level.kroncov: Data Tn is centered before the
estimation.summary.Tenv.ask in plot.Tenv.Gamma list into bat and
square datasets.bic_max to
maxdim in TensEnv_dim, max_iter
to maxiter in manifold1D, epsilon
to tol in ECD, G_ini to
Gamma_init in manifoldFG, G_hat
to Gamma in manifoldFG, Yhat to
pred in PMSE.TRR, TPR as TRR.fit,
TPR,fit.print.Tenv: Prints the call, coefficients from
TPR and TRR, make the output more
concise.print.summary.Tenv: Print call, dimensions of X, Y,
sample size, mse, the coefficient and p_value. (invoked implicitly when
there is no assignment of summary.Tenv).Xn and Yn in
functions: TensEnv_dim, TensPLS_cv2d3d,
Tenv, Tenv_Pval, TPR and
TRR.mvtnorm in DESCRIPTION, it
is not used in the package.Depends in
DESCRIPTION to Imports except for
ManifoldOptim.PMSE,
TensEnv_dim, TensPLS_cv2d3d,
Tenv_Pval.# Usage
> data("bat")
> data("square")TPR, TRR:
Bhat
in the object returned from TPR and TRR is
renamed as coefficients.fitted.values and residuals into
the output.Yn can be vector or
matrix and Xn can be matrix, array or tensor for
TPR, Xn can be vector or matrix and
Yn can be matrix, array or tensor for
TRR.FG_TRR and FG_TPR into
TRR and TPR. Add one more option for
method, “FG”. Also add argument Gamma_init for
“FG” method.Tenv_Pval: Change the data type of outputs from array
to tensor. Change the name se_mat to se.PMSE: Rewrite PMSE. Xn can be
matrix, array or tensor, Yn can be vector or matrix, and
Bhat can be vector, matrix, array or tensor as long as the
dimensions match the ones of Xn and Yn.TRR_sim, TPR_sim: Add two simulation
functions to generate data used in TRR and TPR which can help user
quickly test the functions.Construct S3 object for TRR and TPR with
class attribute “Tenv”
> data("bat")
> fit <- TRR(bat$Xn, bat$Yn, method = "standard")
> class(fit)
> [1] "Tenv"predict.Tenv: Make predictions of new data.summary.Tenv: Append dimensions of X,Y, sample size,
mse, p_val and s.e. to the output object from TPR and
TRR.plot.Tenv: Draw the plot of coefficients from
TRR and TPR, and draw p_val plot from
TRR.vcov.Tenv: No covariance for coefficients. But for
TRR, we print the standard error for coefficients.fitted.default: Calculates the fitted Y for
TPR and TRR separately.residuals.default: Calculate Y minus fitted Y for
TPR and TRRcoef.default: Print the coefficients for
TPR and TRR.FG_TPR, FG_TRR > Help pages for
deprecated functions are available at
help("TRES-deprecated").u=NULL for TPR and
TRR.maxdim=10 and nfold=5 for
TensPLS_cv2d3dPMSE: now accept tensor XnThese 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.
Health stats visible at Monitor.