pkg/R/getData.R

   1 #' @title Acquire data in a clean format
   2 #'
   3 #' @description Take in input data frames and/or files containing raw data, and timezones, and
   4 #'   output a Data object, roughly corresponding to a list where each cell contains all value
   5 #'   for one day (see \code{?Data}).
   6 #'
   7 #' @param ts_data Time-series, as a data frame (DB style: 2 columns, first is date/time,
   8 #'   second is value) or a CSV file
   9 #' @param exo_data Exogenous variables, as a data frame or a CSV file; first comlumn is dates,
  10 #'   next block are measurements for the day, and final block are exogenous forecasts
  11 #' @param input_tz Timezone in the input files ("GMT" or e.g. "Europe/Paris")
  12 #' @param date_format How date/time are stored (e.g. year/month/day hour:minutes;
  13 #'   see \code{strptime})
  14 #' @param working_tz Timezone to work with ("GMT" or e.g. "Europe/Paris")
  15 #' @param predict_at When does the prediction take place ? Integer, in hours. Default: 0
  16 #' @param limit Number of days to extract (default: Inf, for "all")
  17 #'
  18 #' @return A list where data[[i]] contains
  19 #'   \itemize{
  20 #'     \item time: vector of times
  21 #'     \item centered_serie: centered serie
  22 #'     \item level: corresponding level
  23 #'     \item exo: exogenous variables
  24 #'     \item exo_hat: predicted exogenous variables
  25 #'   }
  26 #'
  27 #' @examples
  28 #' ts_data = read.csv(system.file("extdata","pm10_mesures_H_loc.csv",package="talweg"))
  29 #' exo_data = read.csv(system.file("extdata","meteo_extra_noNAs.csv",package="talweg"))
  30 #' data = getData(ts_data, exo_data, limit=120)
  31 #' @export
  32 getData = function(ts_data, exo_data, input_tz="GMT", date_format="%d/%m/%Y %H:%M",
  33     working_tz="GMT", predict_at=0, limit=Inf)
  34 {
  35     # Sanity checks (not full, but sufficient at this stage)
  36     if (!is.character(input_tz) || !is.character(working_tz))
  37         stop("Bad timezone (see ?timezone)")
  38     input_tz = input_tz[1]
  39     working_tz = working_tz[1]
  40     if ( (!is.data.frame(ts_data) && !is.character(ts_data)) ||
  41             (!is.data.frame(exo_data) && !is.character(exo_data)) )
  42         stop("Bad time-series / exogenous input (data frame or CSV file)")
  43     if (is.character(ts_data))
  44         ts_data = ts_data[1]
  45     if (is.character(exo_data))
  46         exo_data = exo_data[1]
  47     predict_at = as.integer(predict_at)[1]
  48     if (predict_at<0 || predict_at>23)
  49         stop("Bad predict_at (0-23)")
  50     if (!is.character(date_format))
  51         stop("Bad date_format (character)")
  52     date_format = date_format[1]
  53
  54     ts_df =
  55         if (is.character(ts_data))
  56             read.csv(ts_data)
  57         else
  58             ts_data
  59     exo_df =
  60         if (is.character(exo_data))
  61             read.csv(exo_data)
  62         else
  63             exo_data
  64     # Convert to the desired timezone (usually "GMT" or "Europe/Paris")
  65     formatted_dates_POSIXlt = strptime(as.character(ts_df[,1]), date_format, tz=input_tz)
  66     ts_df[,1] = format(as.POSIXct(formatted_dates_POSIXlt), tz=working_tz, usetz=TRUE)
  67
  68     line = 1 #index in PM10 file (24 lines for 1 cell)
  69     nb_lines = nrow(ts_df)
  70     nb_exos = ( ncol(exo_df) - 1 ) / 2
  71     data = list()
  72     i = 1 #index of a cell in data
  73     while (line <= nb_lines)
  74     {
  75         time = c()
  76         serie = c()
  77         repeat
  78         {
  79             {
  80                 time = c(time, ts_df[line,1])
  81                 serie = c(serie, ts_df[line,2])
  82                 line = line + 1
  83             };
  84             if (line >= nb_lines + 1 || as.POSIXlt(ts_df[line-1,1])$hour == predict_at)
  85                 break
  86         }
  87
  88         exo = as.data.frame( exo_df[i,2:(1+nb_exos)] )
  89         exo_hat = as.data.frame( exo_df[i,(1+nb_exos+1):(1+2*nb_exos)] )
  90         level = mean(serie, na.rm=TRUE)
  91         centered_serie = serie - level
  92         data[[i]] = list("time"=time, "centered_serie"=centered_serie, "level"=level,
  93             "exo"=exo, "exo_hat"=exo_hat)
  94         if (i >= limit)
  95             break
  96         i = i + 1
  97     }
  98     start = 1
  99     end = length(data)
 100     if (length(data[[1]]$centered_serie) < length(data[[2]]$centered_serie))
 101         start = 2
 102     if (length(data[[length(data)]]$centered_serie) <
 103         length(data[[length(data)-1]]$centered_serie))
 104     {
 105         end = end-1
 106     }
 107     if (start>1 || end<length(data))
 108         data = data[start:end]
 109     data
 110 }