I'm trying to plot and compare weighted histograms and I can't seem find
where to start. I have data similar to this:
Miles2LAX RADAM2005Pct LAWA2005Pct
35.57 .00123 .00684
24.74 .00118 .00187
27.09 .00965 .00876
16.23 .00587 .00397
{ ... 87 rows}
And I'm trying to compare the weighted distribution of "Miles"
using the
two weightings: RADAM2005Pct and LAWA2005Pct. Where do I begin? I'd
ultimately like to have the two distributions on the same plot. E.g.
one cross-hatched right and the other cross-hatched left.
Do I need to pretend this is a survey and the "percentages' are survey
weights? How is such a svydesign formed?
Thanks in advance.
PS I can plot a single unweighted histogram as follows:
>
> SurveyData <- read.table("C:/Data/R/RADAMdata2.csv",
header=TRUE,
sep=",", na.strings="NA", dec=".",
strip.white=TRUE)
> MyX <- c(0,5,10,15,20,30,40,50,60,120)
> hist(SurveyData$Miles2LAX, breaks = MyX,
+ include.lowest = TRUE, right = TRUE,
+ angle = 45,
+ main = paste("Histogram of unweighted distances" ),
+ axes = TRUE, plot = TRUE )
>
>
describe(SurveyData,num.desc=c("mean","median","var","sd","valid.n"),xna
me=NA,maxfac=18,show.pc=TRUE)
Description of SurveyData
Numeric
mean median var sd valid.n
RADAM.TAZ 52.17 51 766.7 27.69 87
Miles2LAX 35.31 32.74 508.1 22.54 87
RADAM2005Vol 1.2e+05 8.199e+04 2.421e+10 1.556e+05 87
RADAM2005Pct 0.01149 0.007853 0.0002221 0.0149 87
LAWA2005Vol 176 76 7.051e+04 265.5 87
LAWA2005Pct 0.01149 0.004963 0.0003007 0.01734 87
Factor
29 Palms Alhambra/San Gab Anaheim
Anaheim/Disneyla Angeles Forest Baldwin Hills/US Bell Cyn
Beverly Hills Big Bear/Arrowhe Brentwood Burbank
Camarillo Canoga Park Castiac Chatsworth
Chino Commerce Compton
RADAM.Name 1 1 1
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
Percent 1.15 1.15 1.15
1.15 1.15 1.15 1.15 1.15
1.15 1.15 1.15 1.15 1.15
1.15 1.15 1.15 1.15 1.15
mode = >1 mode Valid n = 87 87 categories - only first 18 shown
> range(SurveyData$Miles2LAX, na.rm = TRUE)
[1] 2.11 113.99
>
>
Robert Farley
Metro
1 Gateway Plaza
Mail Stop 99-23-7
Los Angeles, CA 90012-2952
Voice: (213)922-2532
Fax: (213)922-2868
www.Metro.net
[[alternative HTML version deleted]]