TSdata: Time Series Example Dataset

Description

Data for Time Series Examples

The data are age-adjusted (2000 U.S. standard) female lung cancer mortality rates (per 100,000 population) for each year from 1996 to 2010.

Usage

TSdata

Arguments

Format

This dataset is an array with dimensions of 51, 15, 4. The rownames of the array are the 51 state and DC abbreviations (2 characters). TSdata[,,1:4] contains the x (time) value, followed by the value for the line, then the lower 95% confidence limit, and finally the upper 95% confidence limit value.

Source

Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov) SEER*Stat Database: Mortality - All COD, Aggregated With State, Total U.S. (1969-2010) (Katrina/Rita Population Adjustment), National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2013. Underlying mortality data provided by NCHS (www.cdc.gov/nchs).

Details

The first dimension [st,,] of 51 elements contains each state or DC. This dimension is referenced by the rownames of the array. The second dimension [,t,] of n elements in this case are the time periods in the time series. Our example uses the years 1996 to 2010 as the time period values. A reasonable number of points is between 200 and 300.

The third dimension [,,v] of 2 or 4 elements is the x or y values during the time period. If the no confidence data is provided, the third dimension is 2:

data[,,1] is the X value
data[,,2] is the mid-Y value (Y)

If a confidence band is being plotted in \var{tsconf} graphs then there are 4 elements.

data[,,1] is the X value
data[,,2] is the mid-Y value (Y)
data[,,3] is the low-Y value
data[..4] is the high-Y value

For example, the x,y coordinates for year=1996 (time period = 1) for the first state (AK) is TSdata[1,1,c(1,2)]. This approach was done to allow a data matrix built for the "tsconf" glyphics to be used for a ts glyphics. This data is used by micromapSEER with the "USStatesDF" border group.

  
  # how to create a new time series data set 
  tempTS <-read.table("...yourfilename.csv",sep=",",header=T)
  yrmat  <-matrix(rep(1996:2010,51),nrow=51,ncol=15,byrow=T)  # year labels
  ratemat<-as.matrix(
         tempTS[,c(8,13,18,23,28,33,38,43,48,53,58,63,68,73,78)]
         )
  locimat<-as.matrix(
         tempTS[,c(9,14,19,24,29,34,39,44,49,54,59,64,69,74,79)]
         )
  hicimat<-as.matrix(
         tempTS[,c(10,15,20,25,30,35,40,45,50,55,60,65,70,75,80)]
         )
  workmat<-cbind(yrmat,ratemat,locimat,hicimat)
  TSdata <-NULL   
  TSdata <-array(workmat,dim=c(51,15,4))
  # change state ab from factors to characters.
  rownames(TSdata)<-as.character(tempTS$stab)