I am working with many signals; each one in a time series but is too many and, I need to make more than 1000 but, I am not sure how to implement it because I not only need the plots but the values of output for each spectrogram stored in a file or an R object. I am sorry I don't have an approach. Can anyone help out, please?
We have a lot of types of plots available in R. And every single time I get a dataset, I have to think for a long time that which type of plot I should use to plot my dataset in order to get information I want (I'm a beginner of R). I don't know whether it's related to my math and stats knowledge or just not familiar with R tech skills. Anybody can tell me the reasons and how to improve that? Thanks many in advance.
I am searching for a method to show progress when doing spatial operations in R. I know there are some for working with raster-data. Does anyone know a way to show progress when doing an operation with vactor-data like union{raster}? It would really help to organize our workflow here.
Thanks
I am trying to create a Control charts with rolling averages in R that closely matches something like this. I am relatively new to R, and my grasp of all the available pages is not great yet.
controlchart with moving average i'd like to recreate
Can someone help me figure this out? Maybe point me to the right packages I should install and study? Any help is appreciated.
I was looking for the answer to creating this type of chart myself.
One solution is the qicharts package. Instead of calling the changes rolling averages or stages (the term Minitab uses), they call the changes breaks.
I saw an interesting chart on engadget today made by Nielsen:
http://www.engadget.com/2011/07/28/nielsen-android-leads-us-smartphone-market-with-39-percent-shar/
original source: http://blog.nielsen.com/nielsenwire/online_mobile/in-u-s-smartphone-market-android-is-top-operating-system-apple-is-top-manufacturer/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+NielsenWire+%28Nielsen+Wire%29
I'd love for someone to replicate it if possible and show the R code. Basic packages or ggplot2 would be great.
I like that the boxes are proportional, that's a key feature :)
Thanks!
You can find several implementations in R under the name of 'mosaic chart'. E.g.:
require("vcd")
data(HairEyeColor)
mosaic(HairEyeColor, shade = TRUE)
Se some examples on e.g. quickR, but searching the R graph gallery is also a good option.
In ggplot2, you can find a sample on learnr's blog.
I have also done some tweaks in ggplot2, please find the attached plot below. It is in Hungarian, but if you are interested, I could clean up the code and post is somewhere.
UPDATE: I have searched for my old script based on comment and uploaded it to pastebin. Sorry, no code clean up and it is quite messy, as I had to make it up for mass reporting from SPSS data files, but I hope you could use it. The usage is simple: load all functions (e.g.: run all lines in R with the source(...) function), and you could generate a mosaic chart of any data frame by specifying two variable names in the parameters of ggMosaicChart(). The plot will be saved to a png file in the working directory (no easy resize in R of the plot as lots of manual tweaks are done to arrange text nicely).
I have translated the strings to English, a basic example (included in the above code) of the mtcars data set:
Count, row- and column percent and also Pearson residuals are shown for each cell.
It's called treemap. R project has packages named "treemap" or "portfolio" for it. Here is how to do: http://flowingdata.com/2010/02/11/an-easy-way-to-make-a-treemap/