tvctomat               package:rmutil               R Documentation

_C_r_e_a_t_e _o_r _a_d_d _t_o _a _t_i_m_e-_v_a_r_y_i_n_g _c_o_v_a_r_i_a_t_e (_t_v_c_o_v) _o_b_j_e_c_t

_D_e_s_c_r_i_p_t_i_o_n:

     `tvctovmat' creates an object of class, tvcov, from a list of
     matrices with time-varying covariates for each individual or one
     matrix or dataframe of such covariate values or combines two such
     objects. It can also add interactions among covariates.

     Such objects can be printed. Methods are available for extracting
     the covariates and their names: `covariates' and `names'. The
     method, `link{transform}', can transform variables in place or by
     adding new variables to the object.

_U_s_a_g_e:

     tvctomat(tvcov, names=NULL, interaction=NULL, ccov=NULL,
             oldtvcov=NULL, dataframe=TRUE)

_A_r_g_u_m_e_n_t_s:

   tvcov: A list of matrices with time-varying covariate values for
          each individual (one column per variable), one matrix or
          dataframe of such covariate values (when there is only one
          such covariate), or an object of class, tvcov. In the first
          two cases, the variables may be factors.

   names: The names of the time-varying covariates in tvcov (if the
          matrices do not have column names) or the names of the
          time-constant covariates for interactions.

interaction: A pair of index numbers or names of variables in tvcov,
          with that class, for which an interaction is to be added or,
          if `ccov' is provided, a set of such names of time-varying
          covariates for creating interactions with the time-constant
          covariates.

    ccov: Time-constant covariates for which an interaction is to be
          introduced with time-varying covariates in tvcov.

oldtvcov: An object of class, tvcov, to which tvcov is to be added.

dataframe: If TRUE and factor variables are present, the covariates are
          stored as a dataframe; if FALSE, they are expanded to
          indicator variables. If no factor variables are present,
          covariates are always stored as a matrix.

_V_a_l_u_e:

     Returns an object of class, tvcov, containing a matrix for the
     covariates (z$tvcov) with one row per response per individual and
     a vector giving the number of observations per individual
     (z$nobs).

_A_u_t_h_o_r(_s):

     J.K. Lindsey

_S_e_e _A_l_s_o:

     `gettvc', `read.list', `restovec', `rmna', `tcctomat',
     `transform'.

_E_x_a_m_p_l_e_s:

     z <- matrix(rpois(20,5),ncol=5)
     print(tvc <- tvctomat(z))
     covariates(tvc)
     names(tvc)
     v <- data.frame(matrix(rep(c("a","b","c","d","e"),4),ncol=5))
     print(tvc2 <- tvctomat(v, oldtvc=tvc))
     covariates(tvc2)
     print(tvc3 <- tvctomat(v, oldtvc=tvc, dataframe=T))
     covariates(tvc3)
     print(tvc4 <- tvctomat(tvc3, interaction=c("z","v")))
     covariates(tvc4)
     x1 <- 1:4
     x2 <- gl(4,1)
     xx <- tcctomat(data.frame(x1,x2))
     tvctomat(tvc3, interaction="z", ccov=xx)
     tvctomat(tvc3, interaction="z", ccov=xx, names="x1")
     tvctomat(tvc3, interaction="z", ccov=xx, names=c("x22","x23","x24"))
     xx <- tcctomat(data.frame(x1,x2), dataframe=T)
     tvctomat(tvc3, interaction="z", ccov=xx)
     tvctomat(tvc3, interaction="z", ccov=xx, names="x1")
     tvctomat(tvc3, interaction="z", ccov=xx, names="x2")

