Description Usage Arguments Value Examples
Fits regularization paths for longitudinal compositional data with grouplasso penalty at a sequence of regularization parameters lambda and fixed degree of freedom of basis.
1 2 3 4 5 6 7 8  FuncompCGL(y, X, Zc = NULL, intercept = TRUE, ref = NULL, k, degree = 3,
basis_fun = c("bs", "OBasis", "fourier"), insert = c("FALSE", "X",
"basis"), method = c("trapezoidal", "step"), interval = c("Original",
"Standard"), Trange, T.name = "TIME", ID.name = "Subject_ID", W = rep(1,
times = p  length(ref)), dfmax = p  length(ref), pfmax = min(dfmax *
1.5, p  length(ref)), lam = NULL, nlam = 100, lambda.factor = ifelse(n
< p1, 0.05, 0.001), tol = 0, mu_ratio = 1.01, outer_maxiter = 1e+08,
outer_eps = 1e08, inner_maxiter = 10000, inner_eps = 1e08)

y 
a vector of response variable. 
X 
a data frame or matrix.

Zc 
A design matrix for control variables, could be missing. Default is NULL. No penalty is imposed. 
intercept 
whether to include intercept. Default is TRUE. 
ref 
reference variable. If 
k 
a scaler, degree of freedom of basis. 
degree 
degree of basis  default value is 3. 
basis_fun 
a function of basis. For now one of the following three types,
Default is 
insert 
way to interpolation. If
Default is 
method 
method used to approximate integral.
Default is 
interval 
a character string sepcifying domain of integral
Default is 
Trange 
range of time points 
T.name, ID.name 
characters specifying names of time varaible and Subject ID respectively in X,
only needed as X is data frame of longitudinal compositinal varaibles.
Default are 
W 
a vector in length of p (the total number of groups), matrix with dimension
Default value is rep(1, times = p). 
dfmax 
limit the maximum number of groups in the model. Useful for very large p, if a partial path is desired  default is p. 
pfmax 
limit the maximum number of groups ever to be nonzero. For example once a group enters the
model along the path, no matter how many times it exits or reenters model through the path,
it will be counted only once. Default is 
lam 
a user supplied lambda sequence. Typically, by leaving this option unspecified users can have the
program compute its own 
nlam 
the length of 
lambda.factor 
the factor for getting the minimal lambda in 
tol 
tolerance for vectors beta'ss to be considered as none zero's. For example, coefficient
β_j for group j, if max(abs(β_j)) < 
mu_ratio 

outer_maxiter 

outer_eps 

inner_maxiter 

inner_eps 

An object with S3 calss FuncompCGL
Z 
integral matrix for longitudinal compositinal predictors with dimension n*(p*k). 
lam 
the actual sequence of 
df 
the number of nonzero groups in estimated coefficients for 
beta 
a matrix of coefficients for 
dim 
dimension of coefficient matrix 
call 
the call that produced this object. 
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33  df_beta = 5
p = 30
beta_C_true = matrix(0, nrow = p, ncol = df_beta)
beta_C_true[3, ] < c(0.8, 0.8 , 0.4 , 1 , 1)
beta_C_true[4, ] < c(0.5, 0.5, 0.6 ,0.6, 0.6)
beta_C_true[1, ] < c(0.5, 0.5, 0.5 , 1, 1)
beta_C_true[2, ] < c(0.8, 0.8, 0.7, 0.6, 0.6)
Data < Model(n = 50, p = p, m = 2, intercept = TRUE,
SNR = 2, sigma = 2,
rho_X = 0, rho_W = 0,
df_W = 5, df_beta = df_beta,
ns = 100, obs_spar = 0.2, theta.add = c(3,4,5),
beta_C = as.vector(t(beta_C_true)))
y < Data$data$y
X < Data$data$Comp
Zc < Data$data$Zc
intercept < Data$data$intercept
k_use < df_beta
m1 < FuncompCGL(y = y, X = X , Zc = Zc, intercept = intercept,
k = k_use, basis_fun = "bs",
insert = "FALSE", method = "t",
dfmax = p, tol = 1e6)
beta < coef(m1, s = m1$lam[20])
#beta < coef(m1)
beta_C < matrix(beta[1:(p*k_use)], nrow = p, byrow = TRUE)
colSums(beta_C)
Non.zero < apply(beta_C, 1, function(x) ifelse(max(abs(x)) == 0, FALSE, TRUE))
Non.zero < (1:p)[Non.zero]
Non.zero
plot(m1, ylab = "L2", p = p , k = k_use)

Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.