Fits hierarchical global polynomial regression model to t-statistics through Newtonian algorithms.
tPoly.newton(tstat, x, df, starts,
pen.order=1,
optim.method = c("nlminb", "BFGS", "CG",
"L-BFGS-B", "Nelder-Mead", "SANN", "NR"),
newton.iter.max = 1500,
scale.conv = 0.001, lfdr.conv = 0.001, NPLL.conv = 0.001,
debugging = FALSE, plotit = TRUE, ...)A numeric vector t-statistics
A numeric matrix of covariates, with nrow(x) being length(tstat)
A numeric scalar or vector of degrees of freedom
An optional numeric vector of starting values. The first element is the r, i.e. log(scale-1). The second parameter is the intercept. The remaining elements are the starting values for the B-spline coefficients (removing the first basis) for each x. When thist argument is not provided, the code starts with a global constant model that is easiest to fit, and then increase the order gradually using the warm starts from lower order fits.
A numeric scalar of the order of derivatives of which squared integration will be used as roughness penalty. Note: The final order of the global polynomial is always pen.order-1.
A character scalar specifying the method of optimization.
A numeric scalar specifying the maximum number of iterations in Newton method.
A small numeric scalar specifying the convergence criterion for the scale parameter.
A small numeric scalar specifying the convergence criterion for the local false discovery rates.
A small numeric scalar specifying the convergence criretion for the negative penalized log likelhood.
A logical scalar. If TRUE, then dump.frame will be called whenever error occurs.
A locgical scalr specifying whether a plot should be generated.
Currently not used.
An list of class hisemit:
A numeric vector of local false discovery rates.
A list of tstat, df and x, which are the same as arguments
A list with
scale.fact: Scale factor
sd.ncp: Equivalent standard deviation of noncentrality parameters
r: A reparameterization of scale.fact
t.cross: sqrt(df*(s^(2/(df+1))-1)/(1-s^(-2*df/(df+1))))
s is the scale.factA numeric vector of mixing proportions for the central t component
A list with
mean: Mean criterion
var: Variance of criterion across observations
grp: Cross-validation group membership
method: The tuning.method used.
final: The minimum mean criterion
A list with
all: All smoothing parameters searched
final: The smoothing parameter used
final.idx: The index of the final spar
A list with
raw: Raw effective number of parameters
logistic: Effective number of parameters after fitting logistic curve as a correction
final: The effective nubmer of parameters in the final model
good.idx: The index of the selected effective number of parameters
A list with
intercept: The fitted intercept
covariate.idx: The index of covariates
f.covariate: Each additive smooth function evaluated at the covariates
f: Fitted smoothing funciton
beta: Estimated regression coefficients
H: Expanded design matrix
asym.vcov: Asymptotic variance-covariance matrix for estimated parameters
A list with
NPLL: Negative penalized log likelihood
logLik: Log likelihood
penalty: Penalty term
saturated.ll: Saturated log likelihood
Long Qu, Dan Nettleton, Jack Dekkers (2012) A hierarchical semiparametric model for incorporating inter-gene relationship information for analysis of genomic data. Biometrics, 68(4):1168-1177
penLik.EMNewton, plot.hisemit, fitted.hisemit, coef.hisemit,
vcov.hisemit, residuals.hisemit, logLik.hisemit,
confint.hisemit, plot.hisemit ,
hisemi-package, pi0-package
# NOT RUN {
<!-- %\dontrun{ -->
# }
# NOT RUN {
# See the example for the package.
# }
# NOT RUN {
<!-- %} -->
# }
Run the code above in your browser using DataLab