(This article was first published on NIRQuimiometría, and kindly contributed to Rbloggers)
When applying SG, we select a moving average window with an odd value “n” for the number of data points. SG fit a polynomial of “p” degree to this data points and give the value to the central point (this is the reason to have an odd value).
We apply also an smooth in the case of “m” = 0, or the first (m=1), second (m=2) or third (m=3) derivatives.
Apply a SavitzkyGolay smoothing filter
Description
Smooth data with a SavitzkyGolay smoothing filter.
Usage
sgolayfilt(x, p = 3, n = p + 3 – p%%2, m = 0, ts = 1)
## S3 method for class ‘sgolayFilter’
filter(filt, x, …)
Arguments
x

signal to be filtered.

p

filter order.

n

filter length (must be odd).

m

return the mth derivative of the filter coefficients.

ts

time scaling factor.

filt

filter characteristics (normally generated by sgolay).

Playing with these three parameter we change the shape of the spectra.
Example:
The upper spectra was taken with p=2, n=7 and m=2
The lower one with p=2, n=31 and m=2
As we can see the size of the window has a big influence on the resolution of the peaks.
How to apply it in R:
(For the lower spectrum)
>sgolay_2_31_2<apply(fattyac$NITm,1,sgolayfilt,p=2,n=31,m=2)
>matplot(wavelengths,sgolay_2_31_2,lty=1,pch=21,
+ xlab=”nm”,ylab=”SG_2_31_2_abs”)
To leave a comment for the author, please follow the link and comment on their blog: NIRQuimiometría.
Rbloggers.com offers daily email updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...