dftorep                package:rmutil                R Documentation

_T_r_a_n_s_f_o_r_m _a _D_a_t_a_f_r_a_m_e _t_o _a _r_e_p_e_a_t_e_d _O_b_j_e_c_t

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

     'dftorep' forms an object of class, 'repeated', from a dataframe
     with the option of removing any observations where response and
     covariate values have NAs. For repeated measurements, observations
     on the same individual must be together in the table. A number of
     validity checks are performed on the data.

     Such objects can be printed and plotted. Methods are available for
     extracting the response, the numbers of observations per
     individual, the times, the weights, the units of
     measurement/Jacobian, the nesting variable, the covariates, and
     their names: 'response', 'nobs', 'times', 'weights', 'delta',
     'nesting', 'covariates', and 'names'.

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

     dftorep(dataframe, response, id=NULL, times=NULL, censor=NULL,
             totals=NULL, weights=NULL, nest=NULL, delta=NULL,
             coordinates=NULL, type=NULL, ccov=NULL, tvcov=NULL, na.rm=T)

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

dataframe: A dataframe.

response: A character vector giving the column name(s) of the dataframe
          for the response variable(s).

      id: A character vector giving the column name of the dataframe
          for the identification numbers of the individuals. If the
          numbers are not consecutive integers, a warning is given.

          If NULL, one observation per individual is assumed if 'times'
          is also NULL, other time series is assumed.

   times: An optional character vector giving the column name of the
          dataframe for the times vector.

  censor: An optional character vector giving the column name(s) of the
          dataframe for the censor indicator(s). This must be the same
          length as 'response'. Responses without censor indicator can
          have a column either of all NAs or all 1s.

  totals: An optional character vector giving the column name(s) of the
          dataframe for the totals for binomial data. This must be the
          same length as 'response'. Responses without censor indicator
          can have a column all NAs.

 weights: An optional character vector giving the column name of the
          dataframe for the weights vector.

    nest: An optional character vector giving the column name of the
          dataframe for the nesting vector within individuals.

          This is the second level of nesting for repeated
          measurements, with the individual being the first level.
          Values for an individual must be consecutive increasing
          integers.

   delta: An optional character vector giving the column name(s) of the
          dataframe for the units of measurement/Jacobian(s) of the
          response(s). This must be the same length as 'response'.
          Responses without units of measurement/Jacobian can have a
          column all NAs.

          If all response variables have the same unit of measurement,
          this can be that one number. If each response variable has
          the same unit of measurement for all its values, this can be
          a numeric vector of length the number of response variables.

coordinates: An optional character vector giving the two or three
          column name(s) of the dataframe for the spatial coordinates.

    type: An optional character vector giving the types of response
          variables: nominal, ordinal, discrete, duration, continuous,
          multivariate, or unknown.

    ccov: An optional character vector giving the column names of the
          dataframe for the time-constant or inter-individual
          covariates. For repeated measurements, if the value is not
          constant for all observations on an individual, an error is
          produced.

   tvcov: An optional character vector giving the column names of the
          dataframe for the time-varying or intra-individual
          covariates.

   na.rm: If TRUE, observations with NAs in any variables selected are
          removed in the object returned. Otherwise, the corresponding
          indicator variable is returned in a slot in the object.

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

     Returns an object of class, 'repeated', containing a list of the
     response object ('z$response', so that, for example, the response
     vector is 'z$response$y'; see 'restovec'), and possibly the two
     classes of covariate objects ('z$ccov' and 'z$tvcov'; see
     'tcctomat' and 'tvctomat').

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

     J.K. Lindsey

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

     'lvna', 'read.list', 'read.rep', 'restovec', 'rmna', 'tcctomat',
     'tvctomat'

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

     y <- data.frame(y1=rpois(20,5),y2=rpois(20,5))
     y[2,2] <- NA
     idd <- c(rep(1,5),rep(2,10),rep(3,5))
     tt <- c(1:5,1:10,1:5)
     totals <- data.frame(tot1=rep(12,20),tot2=rep(12,20))
     x2 <- c(rep(1,5),rep(2,10),rep(3,5))
     df <- data.frame(y,id=idd,tt=tt,totals,x1=rnorm(20),x2=x2)
     df
     dftorep(df,resp=c("y1","y2"),times="tt",id="id",totals=c("tot1","tot2"),
             tvcov="x1",ccov="x2")
     dftorep(df,resp=c("y1","y2"),times="tt",id="id",totals=c("tot1","tot2"),
             tvcov="x1",ccov="x2",na.rm=FALSE)
     # x1 is not a time-constant covariate
     #dftorep(df,resp=c("y1","y2"),times="tt",id="id",ccov="x1",na.rm=FALSE)

