ANN: AverageShiftedHistograms.jl

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

ANN: AverageShiftedHistograms.jl

Josh Day
AverageShiftedHistograms is an implementation of the ASH density estimator.

The ASH estimator is essentially kernel density estimation using a fine-partition histogram.  This package provides support for univariate and bivariate densities, mimicking the functionality of R's ash package.

This package offers several things that R's ash does not.
  1. Cleaner syntax
  2. Update the density estimate with more data:  update!(obj, y).  The advantage over kernel density estimation is that ASH uses O(1) memory.  
  3. Change the smoothing parameter and kernel as opposed to creating a new object: update!(obj, 10, :gaussian).
  4. Approximate summary statistics (mean, var, std, quantile).  Because of the O(1) memory requirement, this is a quick method to get summary statistics from data that may not fit in memory.

Best,
-Josh

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.
Reply | Threaded
Open this post in threaded view
|

Re: ANN: AverageShiftedHistograms.jl

Andreas Noack
Thanks for posting this. I'll definitely take a look.

On Fri, Aug 21, 2015 at 10:48 AM, Josh Day <[hidden email]> wrote:
AverageShiftedHistograms is an implementation of the ASH density estimator.

The ASH estimator is essentially kernel density estimation using a fine-partition histogram.  This package provides support for univariate and bivariate densities, mimicking the functionality of R's ash package.

This package offers several things that R's ash does not.
  1. Cleaner syntax
  2. Update the density estimate with more data:  update!(obj, y).  The advantage over kernel density estimation is that ASH uses O(1) memory.  
  3. Change the smoothing parameter and kernel as opposed to creating a new object: update!(obj, 10, :gaussian).
  4. Approximate summary statistics (mean, var, std, quantile).  Because of the O(1) memory requirement, this is a quick method to get summary statistics from data that may not fit in memory.

Best,
-Josh

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.