[D] What's your favorite concept/rule/theorem in statistics and why?

• Bigsimr.jl

  4 4 8.1 Julia

  Simulate multivariate distributions with arbitrary marginals.

  

I wrote a Julia library that basically applies this idea, but extends it to multivariate distributions. We sample from a multivariate normal, transform the margins to uniform (via the normal cdf), and then transform to the desired distribution using the margins inverse cdf's (called the NORTA algorithm). The caveat is that this transformation is non-linear, so the correlation matrix used to generate the multivariate normal samples is generally not the same as the correlation after transformation. We account for this by numerically solving for the n*(n-1)/2 double integrals to determine what input correlation is necessary to get the desired output correlation. This paper describes the full problem and method for solving.

• r-bigsimr

  2 2 5.5 R

  Simulate arbitrary multivariate distributions

  

We have an R interface to the Julia library, but it still requires Julia to be installed. It uses JuliaCall, and honestly you don't even need to use our R wrapper.

• InfluxDB

  www.influxdata.com featured

  Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.

  

• JuliaCall

  3 260 1.0 HTML

  Embed Julia in R

We have an R interface to the Julia library, but it still requires Julia to be installed. It uses JuliaCall, and honestly you don't even need to use our R wrapper.

Suggest a related project