Kalman filter variants in the closed skew normal setting

The filtering problem (or the dynamic data assimilation problem) is studied for linear and nonlinear systems with continuous state space and over discrete time steps. Filtering approaches based on the conjugate closed skewed normal probability density function are presented. This distribution allows additional flexibility over the usual Gaussian approximations. With linear dynamic systems the filtering problem can be solved in analytical form using expressions for the closed skew normal distribution. With nonlinear dynamic systems an ensemble-based version is proposed for fitting a closed skew normal distribution at each updating step. Numerical examples discuss various special cases of the methods.

[1]  P. Oke,et al.  Implications of the Form of the Ensemble Transformation in the Ensemble Square Root Filters , 2008 .

[2]  A. Jazwinski.  Stochastic Processes and Filtering Theory , 1970 .

[3]  A. Stordal,et al.  Bridging the ensemble Kalman filter and particle filters: the adaptive Gaussian mixture filter , 2011 .

[4]  Henning Omre,et al.  Ensemble Kalman filtering with shrinkage regression techniques , 2011 .

[5]  T. Başar,et al.  A New Approach to Linear Filtering and Prediction Problems , 2001 .

[6]  Tapan Mukerji,et al.  Value of information of seismic amplitude and CSEM resistivity , 2008 .

[7]  G. Evensen.  Data Assimilation: The Ensemble Kalman Filter , 2006 .

[8]  M. Genton,et al.  A skewed Kalman filter , 2005 .

[9]  Geir Evensen,et al.  The Ensemble Kalman Filter: theoretical formulation and practical implementation , 2003 .

[10]  Mohsen Mohammadzadeh,et al.  Bayesian spatial regression models with closed skew normal correlated errors and missing observations , 2012 .

[11]  R. E. Abdel-Aal,et al.  Reservoir property prediction using abductive networks , 2010 .

[12]  Javad Rezaie,et al.  Shrinked (1 − α) ensemble Kalman filter and α Gaussian mixture filter , 2012 .

[13]  Henning Omre,et al.  Bayesian closed-skew Gaussian inversion of seismic AVO data for elastic material properties , 2010 .

[14]  Philippe Naveau,et al.  Estimating the closed skew-normal distribution parameters using weighted moments , 2009 .

[15]  Simon J. Julier,et al.  Skewed approach to filtering , 1998, Defense, Security, and Sensing.

[16]  Timothy J. Robinson,et al.  Sequential Monte Carlo Methods in Practice , 2003 .

[17]  H. Omre,et al.  Bayesian Closed Skew Gaussian Inversion of Seismic AVO Data into Elastic Material Properties , 2009 .

[18]  Arjun K. Gupta,et al.  Additive properties of skew normal random vectors , 2004 .

[19]  Florian Nadel,et al.  Stochastic Processes And Filtering Theory , 2016 .

[20]  P. Embrechts.  Skew-Elliptical Distributions and Their Applications: A Journey Beyond Normality , 2005 .

[21]  T. Mukerji,et al.  The Rock Physics Handbook , 1998 .

[22]  Michael Andrew Christie,et al.  Tenth SPE Comparative Solution Project: a comparison of upscaling techniques , 2001 .

[23]  A. Azzalini,et al.  The multivariate skew-normal distribution , 1996 .

[24]  Geir Evensen,et al.  The ensemble Kalman filter for combined state and parameter estimation: MONTE CARLO TECHNIQUES FOR DATA ASSIMILATION IN LARGE SYSTEMS , 2009 .