0770-0772
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View 0770-0772 as PDF for free.
More details
- Words: 1,324
- Pages: 3
Recent Publications in JSS The following abstracts come from papers recently accepted to the Journal of Statistical Software (JSS). JSS is free; code and papers may be downloaded at no cost. Code means source code, not binaries, so that algorithms may be easily customized or extended. The journal has cross-referencing and a search engine to allow for easy browsing. It can be accessed at www.jstatsoft.org.
JSS EDITORS
Editors-in-Chief Jan de Leeuw, UCLA Achim Zeileis, Wirtschaftsuniversit¨at Wien, Austria
Anestis Antoniadis, U. Joseph Fourier, France Kit Baum, Boston Coll. Roger Bivand, NHH, Norway Adrian Bowman, U. of Glasgow, Scotland Samuel E. Buttrey, Naval Postgraduate School Vince Carey, Harvard U. Nicholas J. Cox, Durham U. Hakan Demirtas, U. of Illinois, Chicago Byron Ellis, AdBrite, Inc. John Fox, McMaster U., Canada Roman Franc¸ois, Mango Solutions Robert Gentleman, Fred Hutchinson Cancer Research Center Paul Gilbert, Bank of Canada Gabor Grothendieck, GKX Associates Inc. Robin K. S. Hankin, Natl. Oceanography Ctr. Wolfgang Hartman, SAS Institute Donald Hedeker, U. Illinois–Chicago Joseph Hilbe, Arizona State U. Kurt Hornik, Wirtschaftsuniversit¨at Wien, Austria Torsten Hothorn, Ludwig-Maximilians-U., M¨unchen, Germany Mortaza Jamshidian, California State U.–Fullerton
Kenneth Knoblauch, Inserm, France Roger Koenker, U. of Illinois Martin M¨achler, ETH Zurich, Switzerland John Maindonald, Australian National U., Australia Patrick Mair, Wirtschaftsuniversit¨at Wien, Austria George Michailidis, U. of Michigan Katharine Mullen, Vrije U., Amsterdam Duncan Murdoch, U. of Western Ontario, Canada Paul Murrell, U. of Auckland, New Zealand Balasubramanian Narasimhan, Stanford U. Erich Neuwirth, U. of Vienna Thomas Petzold, U. of Dresden Colin Rose, tr(I), Australia Deepayan Sarkar, Fred Hutchinson Cancer Research Center R. Woodrow Setzer, U.S. Environmental Protection Agency Arnold J. Stromberg, U. of Kentucky Duncan Temple Lang, UC Davis Luke Tierney, U. of Iowa Antony Unwin, U. of Augsburg, Germany Simon Urbanek, AT&T Labs–Research Pedro Valero-Mora, U. of Val´encia Hadley Wickham, Iowa State U.
c 2008
American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America Journal of Computational and Graphical Statistics, Volume 17, Number 3, Pages 770–772 DOI: 10.1198/106186008X345170 770
R ECENT P UBLICATIONS IN JSS
771
An Implementation of Bayesian Adaptive Regression Splines (BARS) in C with S and R Wrappers, Garrick Wallstrom, Jeffrey Liebner, and Robert E. Kass, Journal of Statistical Software, http:// www.jstatsoft.org/ v26/ i01.
BARS (DiMatteo, Genovese, and Kass 2001) uses the powerful reversible-jump MCMC engine to perform spline-based generalized nonparametric regression. It has been shown to work well in terms of having small mean-squared error in many examples (smaller than known competitors), as well as producing visually appealing fits that are smooth (filtering out high-frequency noise) while adapting to sudden changes (retaining high-frequency signal). However, BARS is computationally intensive. The original implementation in S was too slow to be practical in certain situations, and was found to handle some data sets incorrectly. We have implemented BARS in C for the normal and Poisson cases, the latter being important in neurophysiological and other point-process applications. The C implementation includes all needed subroutines for fitting Poisson regression, manipulating B-splines (using code created by Bates and Venables), and finding starting values for Poisson regression (using code for density estimation created by Kooperberg). The code utilizes only freely-available external libraries (LAPACK and BLAS) and is otherwise self-contained. We have also provided wrappers so that BARS can be used easily within S or R. Key Words: curve-fitting, free-knot splines, nonparametric regression, peri-stimulus time histogram, Poisson process.
Rational Arithmetic Mathematica Functions to Evaluate the Two-Sided One Sample K-S Cumulative Sampling Distribution, J. Randall Brown and Milton E. Harvey, Journal of Statistical Software, http:// www.jstatsoft.org/ v26/ i02. One of the most widely used goodness-of-fit tests is the two-sided one sample Kolmogorov-Smirnov (K-S) test which has been implemented by many computer statistical software packages. To calculate a two-sided p-value (evaluate the cumulative sampling distribution), these packages use various methods including recursion formulae, limiting distributions, and approximations of unknown accuracy developed over thirty years ago. Based on an extensive literature search for the two-sided one sample K-S test, this paper identifies an exact formula for sample sizes up to 31, six recursion formulae, and one matrix formula that can be used to calculate a p-value. To ensure accurate calculation by avoiding catastrophic cancelation and eliminating rounding error, each of these formulae is implemented in rational arithmetic. For the six recursion formulae and the matrix formula, computational experience for sample sizes up to 500 shows that computational times are increasing functions of both the sample size and the number of digits in the numerator and denominator integers of the rational number test statistic. The computational times of the seven formulae vary immensely but the Durbin recursion formula is almost always the fastest. Linear search is used to calculate the inverse of the cumulative sampling distribution (find the confidence interval half-width) and tables of calculated half-widths are presented for sample sizes up to 500. Using calculated half-widths as input, computational times for the fastest formula, the Durbin recursion formula, are given for sample sizes up to two thousand. Key Words: K-S cumulative sampling distributions, K-S two-sided one sample probabilities, K-S confidence bands, rational arithmetic.
772
R ECENT P UBLICATIONS IN JSS
Arbitrary Precision Mathematica Functions to Evaluate the One-Sided One Sample K-S Cumulative Sampling Distribution, J. Randall Brown and Milton E. Harvey, Journal of Statistical Software, http:// www.jstatsoft.org/ v26/ i03.
Efficient rational arithmetic methods that can exactly evaluate the cumulative sampling distribution of the one-sided one sample Kolmogorov-Smirnov (K-S) test have been developed by Brown and Harvey (2007) for sample sizes n up to fifty thousand. This paper implements in arbitrary precision the same 13 formulae to evaluate the one-sided one sample K-S cumulative sampling distribution. Computational experience identifies the fastest implementation which is then used to calculate confidence interval bandwidths and p values for sample sizes up to ten million. Key Words: K-S cumulative sampling distributions, K-S one-sided one sample probabilities, K-S confidence bands, arbitrary precision arithmetic.
An R Package for a General Class of Inverse Gaussian Distributions, V´ıctor Leiva, Hugo Hern´andez, and Antonio Sanhueza, Journal of Statistical Software, http:// www. jstatsoft.org/ v26/ i04. The inverse Gaussian distribution is a positively skewed probability model that has received great attention in the last 20 years. Recently, a family that generalizes this model called inverse Gaussian type distributions has been developed. The new R package named ig has been designed to analyze data from inverse Gaussian type distributions. This package contains basic probabilistic functions, lifetime indicators and a random number generator from this model. Also, parameter estimates and diagnostics analysis can be obtained using likelihood methods by means of this package. In addition, goodness-of-fit methods are implemented in order to detect the suitability of the model to the data. The capabilities and features of the ig package are illustrated using simulated and real data sets. Furthermore, some new results related to the inverse Gaussian type distribution are also obtained. Moreover, a simulation study is conducted for evaluating the estimation method implemented in the ig package. Key Words: diagnostics, goodness of fit, lifetime analysis, likelihood methods.
Model Averaging Software for Dichotomous Dose Response Risk Estimation, Matthew W. Wheeler and A. John Bailer, Journal of Statistical Software, http:// www.jstatsoft. org/ v26/ i05. Model averaging has been shown to be a useful method for incorporating model uncertainty in quantitative risk estimation. In certain circumstances this technique is computationally complex, requiring sophisticated software to carry out the computation. We introduce software that implements model averaging for risk assessment based upon dichotomous dose-response data. This software, which we call Model Averaging for Dichotomous Response Benchmark Dose (MADr-BMD), fits the quantal response models, which are also used in the U.S. Environmental Protection Agency benchmark dose software suite, and generates a model-averaged dose response model to generate benchmark dose and benchmark dose lower bound estimates. The software fulfills a need for risk assessors, allowing them to go beyond one single model in their risk assessments based on quantal data by focusing on a set of models that describes the experimental data. Key Words: bootstrapping, information criteria, model uncertainty.
JSS EDITORS
Editors-in-Chief Jan de Leeuw, UCLA Achim Zeileis, Wirtschaftsuniversit¨at Wien, Austria
Anestis Antoniadis, U. Joseph Fourier, France Kit Baum, Boston Coll. Roger Bivand, NHH, Norway Adrian Bowman, U. of Glasgow, Scotland Samuel E. Buttrey, Naval Postgraduate School Vince Carey, Harvard U. Nicholas J. Cox, Durham U. Hakan Demirtas, U. of Illinois, Chicago Byron Ellis, AdBrite, Inc. John Fox, McMaster U., Canada Roman Franc¸ois, Mango Solutions Robert Gentleman, Fred Hutchinson Cancer Research Center Paul Gilbert, Bank of Canada Gabor Grothendieck, GKX Associates Inc. Robin K. S. Hankin, Natl. Oceanography Ctr. Wolfgang Hartman, SAS Institute Donald Hedeker, U. Illinois–Chicago Joseph Hilbe, Arizona State U. Kurt Hornik, Wirtschaftsuniversit¨at Wien, Austria Torsten Hothorn, Ludwig-Maximilians-U., M¨unchen, Germany Mortaza Jamshidian, California State U.–Fullerton
Kenneth Knoblauch, Inserm, France Roger Koenker, U. of Illinois Martin M¨achler, ETH Zurich, Switzerland John Maindonald, Australian National U., Australia Patrick Mair, Wirtschaftsuniversit¨at Wien, Austria George Michailidis, U. of Michigan Katharine Mullen, Vrije U., Amsterdam Duncan Murdoch, U. of Western Ontario, Canada Paul Murrell, U. of Auckland, New Zealand Balasubramanian Narasimhan, Stanford U. Erich Neuwirth, U. of Vienna Thomas Petzold, U. of Dresden Colin Rose, tr(I), Australia Deepayan Sarkar, Fred Hutchinson Cancer Research Center R. Woodrow Setzer, U.S. Environmental Protection Agency Arnold J. Stromberg, U. of Kentucky Duncan Temple Lang, UC Davis Luke Tierney, U. of Iowa Antony Unwin, U. of Augsburg, Germany Simon Urbanek, AT&T Labs–Research Pedro Valero-Mora, U. of Val´encia Hadley Wickham, Iowa State U.
c 2008
American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America Journal of Computational and Graphical Statistics, Volume 17, Number 3, Pages 770–772 DOI: 10.1198/106186008X345170 770
R ECENT P UBLICATIONS IN JSS
771
An Implementation of Bayesian Adaptive Regression Splines (BARS) in C with S and R Wrappers, Garrick Wallstrom, Jeffrey Liebner, and Robert E. Kass, Journal of Statistical Software, http:// www.jstatsoft.org/ v26/ i01.
BARS (DiMatteo, Genovese, and Kass 2001) uses the powerful reversible-jump MCMC engine to perform spline-based generalized nonparametric regression. It has been shown to work well in terms of having small mean-squared error in many examples (smaller than known competitors), as well as producing visually appealing fits that are smooth (filtering out high-frequency noise) while adapting to sudden changes (retaining high-frequency signal). However, BARS is computationally intensive. The original implementation in S was too slow to be practical in certain situations, and was found to handle some data sets incorrectly. We have implemented BARS in C for the normal and Poisson cases, the latter being important in neurophysiological and other point-process applications. The C implementation includes all needed subroutines for fitting Poisson regression, manipulating B-splines (using code created by Bates and Venables), and finding starting values for Poisson regression (using code for density estimation created by Kooperberg). The code utilizes only freely-available external libraries (LAPACK and BLAS) and is otherwise self-contained. We have also provided wrappers so that BARS can be used easily within S or R. Key Words: curve-fitting, free-knot splines, nonparametric regression, peri-stimulus time histogram, Poisson process.
Rational Arithmetic Mathematica Functions to Evaluate the Two-Sided One Sample K-S Cumulative Sampling Distribution, J. Randall Brown and Milton E. Harvey, Journal of Statistical Software, http:// www.jstatsoft.org/ v26/ i02. One of the most widely used goodness-of-fit tests is the two-sided one sample Kolmogorov-Smirnov (K-S) test which has been implemented by many computer statistical software packages. To calculate a two-sided p-value (evaluate the cumulative sampling distribution), these packages use various methods including recursion formulae, limiting distributions, and approximations of unknown accuracy developed over thirty years ago. Based on an extensive literature search for the two-sided one sample K-S test, this paper identifies an exact formula for sample sizes up to 31, six recursion formulae, and one matrix formula that can be used to calculate a p-value. To ensure accurate calculation by avoiding catastrophic cancelation and eliminating rounding error, each of these formulae is implemented in rational arithmetic. For the six recursion formulae and the matrix formula, computational experience for sample sizes up to 500 shows that computational times are increasing functions of both the sample size and the number of digits in the numerator and denominator integers of the rational number test statistic. The computational times of the seven formulae vary immensely but the Durbin recursion formula is almost always the fastest. Linear search is used to calculate the inverse of the cumulative sampling distribution (find the confidence interval half-width) and tables of calculated half-widths are presented for sample sizes up to 500. Using calculated half-widths as input, computational times for the fastest formula, the Durbin recursion formula, are given for sample sizes up to two thousand. Key Words: K-S cumulative sampling distributions, K-S two-sided one sample probabilities, K-S confidence bands, rational arithmetic.
772
R ECENT P UBLICATIONS IN JSS
Arbitrary Precision Mathematica Functions to Evaluate the One-Sided One Sample K-S Cumulative Sampling Distribution, J. Randall Brown and Milton E. Harvey, Journal of Statistical Software, http:// www.jstatsoft.org/ v26/ i03.
Efficient rational arithmetic methods that can exactly evaluate the cumulative sampling distribution of the one-sided one sample Kolmogorov-Smirnov (K-S) test have been developed by Brown and Harvey (2007) for sample sizes n up to fifty thousand. This paper implements in arbitrary precision the same 13 formulae to evaluate the one-sided one sample K-S cumulative sampling distribution. Computational experience identifies the fastest implementation which is then used to calculate confidence interval bandwidths and p values for sample sizes up to ten million. Key Words: K-S cumulative sampling distributions, K-S one-sided one sample probabilities, K-S confidence bands, arbitrary precision arithmetic.
An R Package for a General Class of Inverse Gaussian Distributions, V´ıctor Leiva, Hugo Hern´andez, and Antonio Sanhueza, Journal of Statistical Software, http:// www. jstatsoft.org/ v26/ i04. The inverse Gaussian distribution is a positively skewed probability model that has received great attention in the last 20 years. Recently, a family that generalizes this model called inverse Gaussian type distributions has been developed. The new R package named ig has been designed to analyze data from inverse Gaussian type distributions. This package contains basic probabilistic functions, lifetime indicators and a random number generator from this model. Also, parameter estimates and diagnostics analysis can be obtained using likelihood methods by means of this package. In addition, goodness-of-fit methods are implemented in order to detect the suitability of the model to the data. The capabilities and features of the ig package are illustrated using simulated and real data sets. Furthermore, some new results related to the inverse Gaussian type distribution are also obtained. Moreover, a simulation study is conducted for evaluating the estimation method implemented in the ig package. Key Words: diagnostics, goodness of fit, lifetime analysis, likelihood methods.
Model Averaging Software for Dichotomous Dose Response Risk Estimation, Matthew W. Wheeler and A. John Bailer, Journal of Statistical Software, http:// www.jstatsoft. org/ v26/ i05. Model averaging has been shown to be a useful method for incorporating model uncertainty in quantitative risk estimation. In certain circumstances this technique is computationally complex, requiring sophisticated software to carry out the computation. We introduce software that implements model averaging for risk assessment based upon dichotomous dose-response data. This software, which we call Model Averaging for Dichotomous Response Benchmark Dose (MADr-BMD), fits the quantal response models, which are also used in the U.S. Environmental Protection Agency benchmark dose software suite, and generates a model-averaged dose response model to generate benchmark dose and benchmark dose lower bound estimates. The software fulfills a need for risk assessors, allowing them to go beyond one single model in their risk assessments based on quantal data by focusing on a set of models that describes the experimental data. Key Words: bootstrapping, information criteria, model uncertainty.
