Finding the function of the most likely distribution when using sm.density.compare
I am using the "sm" package for studying distributions in my datasets. For those curious I am looking at recruitment practices as a function of age and trying to identify if the age distribution change for groups change due to attributes such as gender or education.
The SM package is a new experience for me and I am trying to find a function describing the most likely density function generating the dataset provided that the null hypotheis is true (that both density plots are generated by random samples taken from the same distribution). I do not have the reputation necessary to post images but I found this one on Imgur that also is generated with sm.density.compare.
What we see in the image is two kernel density plots and a teal area that I understand to be the reference band containing the 95 % likelihood of containing the true density plot provided that the two lines have been generated by data coming from the same distribution.
What I would like to find out is a way to calculate a vector going through the most likely point within the reference band for each value on the x-axis. As per the rules I am of course open to suggestions that I am either crazy or should use another package.
Apparently there is some bug in either Rstudio or R that prevented me from solving the problem in the most obvious way. As everything else in R sm.density.compare generates an object but for me it failed when I tried to save it.
The problem automagically went away when I first plotted the object (default behaviour when sm.density.compare is executed) and then did the same thing but also creating a new object.
So:
sm.density.compare(a, b, model="equal", xlab=Year of Birth")
DensityObject = sm.density.compare(a, b, model="equal", xlab=Year of Birth")
And then the actual line can be calculated by:
DensityObject$average = rowMeans(cbind(DensityObject$upper, DensityObject$lower), na.rm=TRUE)
- Assign groups in dataframe using vectors containing start and end indexes
- Is there a product operator (or work around) in SQLite?
- Loading Magick package after tidyverse
- Make silhouette icons in tmap
- Ignoring NA cases when getting column index of lowest value in row
- plm object to LaTeX table
- Why is my bslib theme not coloring the navbar as expected?
- Adding a footnote to a single row label in a gtsummary table
- How to order a data frame by one descending and one ascending column?
- dplyr is turning dates to doubles when mutating dataframe
- How to display headers from a quarto with a for loop and plotly graphs?
- mixed date converted from default mm/dd/yy in excel to dd/mm/yy
- Git lfs version https://git-lfs.github.com/spec/v1 oid error
- Generate list of lead values
- How to find the date a categorical variable was last active?
- R - could not find function "cld"
- Overlay sf object on ggmap - not aligned, specific to region
- Using ggplot and alleffects in tandem
- Assign categories from JSON column in data table
- Opposite of tidyr::separate, concatenating multiple columns into one
- How to Stack Painted Phylogenetic Trees in R Like ggdensitree but with Colored Regimes
- Counting number of viable pathways in a network diagram from a specific node input
- Report p-value from lme model (random intercept) with gtsummary using tbl_summary() and add_p()
- dynamically remove all nav panel from tabsetPanel
- If x and y have the same length, do qqplot(x,y) and plot(sort(x),sort(y)) yield the same plot in R?
- Applying SPEI::hargreaves function to time series from each pixel of SpatRaster using terra package R?
- Turn dataframe with frequency into a frequency table for chi-square test
- Print result of a function in R
- How to require user authentication in R Shiny before users see any part of the app using shinyauthr?
- Determine foreground / background colour in bslib app with dark/light mode