R/timeserplot.r

In datana: Datasets and Functions to Accompany Analisis De Datos Con R

#' Produces a time series plot, of variable 'y' as a function of 'x' by
#' an observational unit factor.
#'
#'   Both 'y' and 'x' must be numeric variables, and the column representing
#'   the observational unit, must be
#'   a factor. This factor identifies the longitudinal context of the data,
#'    for instance, a student
#'   being measured on time. Besides, two more factors can be added to the
#'   plotting details, in order
#'    to represent the potential variability among them.
#'
#'
#' @title Produces a time series plot
#' @param data a dataframe with at least tree columns representing the
#' response variable ("y"),
#'   the main predictor variable ("x"), and a variable indicating the
#'   observational unit ("obs.unit").
#' @param y a character giving the column name of the response variable
#'  or variable of interest.
#' @param x a character giving the column name of the main predictor
#' variable. Generally this variable
#' is time.
#' @param obs.unit a character giving the column name containing the info
#' of the observational unit.
#' @param factor1 an optional character having the name of a column having
#' a factor variable (e.g., treatment). The
#'  detault value is set to NULL.
#' @param factor2 an optional character having the name of a column having
#' another factor variable (e.g., species). The
#'  detault value is set to NULL.
#' @param only.lines a logic value if only lines, but not including dots,
#'  are going to be drwan in the plot. The detault
#'  value is set to FALSE.
#' @param xlab Label for the X-axis
#' @param ylab Label for the Y-axis
#' @param linetype.lab is an optional string to be used as the title of the
#'  factor being represented by lines. It is only
#'  needed if factor1 and factor2 are defined. See example.
#' @param factor2.line a logic value if the second factor, factor2, is going
#'  to be segregated according to the
#'  type of lines. The detault  value is set to TRUE.
#' @param factor2.col a logic value if the second factor, factor2, is going to
#'  be segregated according to the
#'  color of the lines only. The detault  value is set to FALSE.
#' @param max.y.all A number representing the maximum level of Y-axis for all
#' classes
#' @param levels.i.want A vector having the levels for the factor under study
#' @param col.lev.i.want A vector having the colors to be used for the factor
#'  under study
#' @param col.lines A string specifying the single color to be used for the
#'  lines of the timeseries
#'
#' @return This function returns a time series plot
#' @author Christian Salas-Eljatib
#' @references
#' Salas-Eljatib, C. 2021. Análisis de datos con el programa estadístico R:
#' una introducción
#'  aplicada. Ediciones Universidad Mayor, Santiago, Chile. 170 p.
#'   \url{https://eljatib.com/rlibro}
#' @note Please, uses the function with caution, and run first the examples to
#'  understand it better.
#' @examples
#' data(ficdiamgr, package="datana")
#' df <- ficdiamgr
#' head(df)
#' str(df)
#' df$site<-as.factor(df$site)
#' df$species<-as.factor(df$species)
#' table(df$tree,df$species)
#' table(df$species,df$site)
#'#
#' timeserplot(df, y="dbh", x="time", obs.unit = "tree")
#' timeserplot(df, y="dbh", x="time", obs.unit = "tree", only.lines = TRUE)
#'#
#'## Otros ejemplos de uso de la funcion
#'timeserplot(df, y="dbh", x="time", obs.unit = "tree", col.lines = "blue",
#' only.lines = TRUE)
#' timeserplot(df, y="dbh", x="time", obs.unit = "tree", only.lines = FALSE)
#'#
#'timeserplot(df, y="dbh", x="time", obs.unit = "tree", factor1="site")
#'timeserplot(df, y="dbh", x="time", obs.unit = "tree", factor1="site",
#' factor2= "species")
#' timeserplot(df, y="dbh", x="time", obs.unit = "tree", factor1="site",
#'  factor2= "species", factor2.col = TRUE, only.lines = TRUE)
#' @rdname timeserplot
#' @export
#'
#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
timeserplot <- function(data=data, y=y, x=x, obs.unit=obs.unit,
                        factor1 = NA, factor2=NA,only.lines=FALSE,
                        ylab=NA, xlab=NA,linetype.lab=NA,
                factor2.line=TRUE,factor2.col=FALSE,col.lines="black",
                max.y.all=NA,levels.i.want=FALSE, col.lev.i.want=FALSE){
  db<-data;

  if(prod(is.na(factor1))){has.factor1 <-0} else {has.factor1 <-1}
  if(prod(is.na(factor2))){has.factor2 <-0} else {has.factor2 <-1}

  if(has.factor1==0 & has.factor2==0){
    df.h <- data.frame(assign(obs.unit,db[,obs.unit]),
                       assign(x,db[,x]),assign(y,db[,y]))
    names(df.h) <- c(obs.unit,x,y) } else {nada=111}
  if(has.factor1==1 & has.factor2==0){
    df.h <- data.frame(assign(obs.unit,db[,obs.unit]),
          assign(factor1,db[,factor1]),assign(x,db[, x]),assign(y,db[,y]))
    names(df.h) <- c(obs.unit,factor1,x,y) } else {nada=111}
  if(has.factor1==1 & has.factor2==1){
    df.h <- data.frame(assign(obs.unit,db[,obs.unit]),
  assign(factor1,db[,factor1]),assign(factor2,db[,factor2]),assign(x,db[, x]),
    assign(y,db[,y]))
    names(df.h) <- c(obs.unit,factor1,factor2,x,y) } else {nada=111}

  if(prod(is.na(xlab))){xlab.h<-x} else {xlab.h <- xlab}
  if(prod(is.na(ylab))){ylab.h<-y} else {ylab.h <- ylab}
  if(prod(is.na(linetype.lab))){
    linetype.lab.h<-factor2} else {linetype.lab.h <- linetype.lab}

  if(has.factor1==0 & has.factor2==0 & only.lines==FALSE){
  ggplot2::ggplot(df.h, ggplot2::aes(x=df.h[,x], y=df.h[,y],
                                     group=df.h[,obs.unit]))
    p<- ggplot2::ggplot(df.h, ggplot2::aes(x=df.h[,x],
              y=df.h[,y], group=df.h[,obs.unit])) +
      ggplot2::geom_line()+ ggplot2::geom_point()+
      ggplot2::xlab(xlab.h) + ggplot2::ylab(ylab.h)
    p <- p +
      ggplot2::theme(panel.background = ggplot2::element_blank(),
                     panel.border = ggplot2::element_blank(),
                     panel.grid.major = ggplot2::element_blank(),
                     panel.grid.minor = ggplot2::element_blank(),
                     axis.line = ggplot2::element_line(colour = "black"))

    print(p)
  } else {nada=0}

  if(has.factor1==0 & has.factor2==0 & only.lines==TRUE){
    ggplot2::ggplot(df.h, ggplot2::aes(x=df.h[,x], y=df.h[,y],
                                       group=df.h[,obs.unit]))
    p<- ggplot2::ggplot(df.h, ggplot2::aes(x=df.h[,x], y=df.h[,y],
                                           group=df.h[,obs.unit])) +
      ggplot2::geom_line()+
      ggplot2::xlab(xlab.h) + ggplot2::ylab(ylab.h)
    p <- p +
      ggplot2::theme(panel.background = ggplot2::element_blank(),
                     panel.border = ggplot2::element_blank(),
                     panel.grid.major = ggplot2::element_blank(),
                     panel.grid.minor = ggplot2::element_blank(),
                     axis.line = ggplot2::element_line(colour = "black"))
    print(p)
  } else {nada=0}

  ##for one factor plotting
  if(has.factor1==1 & has.factor2==0  & only.lines==FALSE){
    p<-  ggplot2::qplot(y=df.h[,y],x=df.h[,x], group=df.h[,obs.unit],
facets=~df.h[,factor1], data=df.h,geom=c("point", "line"),
ylab=ylab.h,xlab=xlab.h)
    p <- p +
      ggplot2::theme(panel.background = ggplot2::element_blank(),
                     panel.border = ggplot2::element_blank(),
                     panel.grid.major = ggplot2::element_blank(),
                     panel.grid.minor = ggplot2::element_blank(),
                     axis.line = ggplot2::element_line(colour = "black"))
    print(p)
  } else {nada=0}

  if(has.factor1==1 & has.factor2==0  & only.lines==TRUE){
    p<-  ggplot2::qplot(y=df.h[,y],x=df.h[,x], group=df.h[,obs.unit],
                        facets=~df.h[,factor1], data=df.h,geom=c("line"),
                        ylab=ylab.h,xlab=xlab.h)
    p <- p +
      ggplot2::theme(panel.background = ggplot2::element_blank(),
                     panel.border = ggplot2::element_blank(),
                     panel.grid.major = ggplot2::element_blank(),
                     panel.grid.minor = ggplot2::element_blank(),
                     axis.line = ggplot2::element_line(colour = "black"))
    print(p)
  } else {nada=0}

  ##for two factors plotting
if(has.factor1==1 & has.factor2==1 & factor2.col==TRUE){
  factor2.line=FALSE} else {nada=999}

if(has.factor1==1 & has.factor2==1 & factor2.line==TRUE  & only.lines==FALSE){
    p<-  ggplot2::ggplot(df.h, ggplot2::aes(x=df.h[,x], y=df.h[,y],
                                            group=df.h[,obs.unit])) +
      ggplot2::geom_line(ggplot2::aes(linetype=df.h[,factor2]))+
      ggplot2::facet_wrap(df.h[,factor1])+
      ggplot2::geom_point()+
      ggplot2::xlab(xlab.h) + ggplot2::ylab(ylab.h) +
      ggplot2::labs(linetype=linetype.lab.h)
    p <- p +
      ggplot2::theme(panel.background = ggplot2::element_blank(),
                     panel.border = ggplot2::element_blank(),
                     panel.grid.major = ggplot2::element_blank(),
                     panel.grid.minor = ggplot2::element_blank(),
                     axis.line = ggplot2::element_line(colour = "black"))
    print(p)
  } else {nada=0}

  if(has.factor1==1 & has.factor2==1 & factor2.line==TRUE  & only.lines==TRUE){
    p<-  ggplot2::ggplot(df.h, ggplot2::aes(x=df.h[,x], y=df.h[,y],
                                            group=df.h[,obs.unit])) +
      ggplot2::geom_line(ggplot2::aes(linetype=df.h[,factor2]))+
      ggplot2::facet_wrap(df.h[,factor1])+
      ggplot2::xlab(xlab.h) + ggplot2::ylab(ylab.h) +
      ggplot2::labs(linetype=linetype.lab.h)
    p <- p +
      ggplot2::theme(panel.background = ggplot2::element_blank(),
                     panel.border = ggplot2::element_blank(),
                     panel.grid.major = ggplot2::element_blank(),
                     panel.grid.minor = ggplot2::element_blank(),
                     axis.line = ggplot2::element_line(colour = "black"))
    print(p)
  } else {nada=0}

  if(has.factor1==1 & has.factor2==1 & factor2.col==TRUE  & only.lines==FALSE){
    p<-  ggplot2::ggplot(df.h, ggplot2::aes(x=df.h[,x], y=df.h[,y],
                                            group=df.h[,obs.unit])) +
      ggplot2::geom_line(ggplot2::aes(col=df.h[,factor2]))+
      ggplot2::facet_wrap(df.h[,factor1])+
      ggplot2::geom_point()+
      ggplot2::xlab(xlab.h) + ggplot2::ylab(ylab.h) +
      ggplot2::labs(col=linetype.lab.h)
    p <- p +
      ggplot2::theme(panel.background = ggplot2::element_blank(),
                     panel.border = ggplot2::element_blank(),
                     panel.grid.major = ggplot2::element_blank(),
                     panel.grid.minor = ggplot2::element_blank(),
                     axis.line = ggplot2::element_line(colour = "black"))
    print(p)
  } else {nada=0}

  if(has.factor1==1 & has.factor2==1 & factor2.col==TRUE  & only.lines==TRUE){
    p<-  ggplot2::ggplot(df.h, ggplot2::aes(x=df.h[,x], y=df.h[,y],
                                            group=df.h[,obs.unit])) +
      ggplot2::geom_line(ggplot2::aes(col=df.h[,factor2]))+
      ggplot2::facet_wrap(df.h[,factor1])+
      ggplot2::xlab(xlab.h) + ggplot2::ylab(ylab.h) +
      ggplot2::labs(col=linetype.lab.h)
    p <- p +
      ggplot2::theme(panel.background = ggplot2::element_blank(),
                     panel.border = ggplot2::element_blank(),
                     panel.grid.major = ggplot2::element_blank(),
                     panel.grid.minor = ggplot2::element_blank(),
                     axis.line = ggplot2::element_line(colour = "black"))
    print(p)
  } else {nada=0}

}
#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@