this is for holding javascript data
Jack O'Brien edited untitled.tex
almost 8 years ago
Commit id: 252b9213caef7132f3975f68b143f3ffa8725d70
deletions | additions
diff --git a/untitled.tex b/untitled.tex
index ab1f4cc..949ee64 100644
--- a/untitled.tex
+++ b/untitled.tex
...
\section{Fitting CARMA Models}
\subsection{CARMA_Pack} \subsection{CARMA\_Pack}
Computationally analyzing and fitting CARMA models requires a fairly sophisticated package. Though many scripting and data analysis languages offer libraries for ARMA analysis such as R and Matlab, few offer tools for the analysis of continuous-time data sets. For our analysis, we have decided to use a python library written in C++ called
carma_pack. Carma_pack carma\_pack. Carma\_pack was originally written for the purpose of analyzing astronomical time-series, specifically, AGN. It has been used successfully to derive the parametric structure of light curves of AGN taken by the Kepler satellite, and has shown to be a useful tool in the analysis of SDSS light curves (Kelly et. al 2014). This package allows us to easily and conveniently fit CARMA process to light curves, determine the optimal model orders, and perform basic stochastic analysis of our best fit models.
Carma_pack Carma\_pack fits CARMA processes to light curves using the Metropolis-Hastings Markov-chain Monte-Carlo method (citation needed) with a Kalman filter providing the likelihood estimations for the proposals.
\subsection{Model Orders}