Journal of Computational & Graphical Statistics articles from March 2000

Journal of Computational & Graphical Statistics archives from March 2000

Optimization Transfer Using Surrogate Objective Functions.
March 1, 2000... The well-known EM algorithm is art optimization transfer algorithm that depends on the notion of incomplete or missing data. By invoking convexity arguments, one can construct a variety of other optimization transfer algorithms that do not...

Discussion.(response to article by Kenneth Lange, et al., in this issue p. 1.)
March 1, 2000... 1. GENERAL COMMENTS The article describes the idea of using a "transfer" function for optimizing functions that are combinations of convex functions. The idea is the same as that of iterative majorization, which has been used for more than...

Discussion.(response to article by Kenneth Lange et al., in this issue p. 1.)
March 1, 2000... 1. INTRODUCTION It is a pleasure to comment on such a well-written and obviously important article. We agree with the basic explicit message of Lange, Hunter, and Yang (LHY). Their so-called "optimization transfer" algorithms form a...

Discussion.(response to article by Kenneth Lange, et al., in this issue p. 1.)
March 1, 2000... I have learned a great deal from reading this article, and I believe that many people whose research involves scientific computing will benefit from it. In the following, I will discuss several related algorithms and an open question. 1. A...

Discussion.(response to article by Kenneth Lange, et al., in this issue p. 1.)
March 1, 2000... 1. IT'S ALL IN THE NAME! Of the several reasons for the popularity of the EM algorithm after the publication of Dempster, Laird, and Rubin (1977), one is its name. Almost at the instant of inquiring what EM stands for, the curious mind is...

Rejoinder.(response to articles in this issue pp. 21, 26, 32, 35)
March 1, 2000... 1. WHAT'S IN A NAME? We thank the discussants for their insightful and substantive comments. In replying, we open with the least substantive issue--the name of the algorithm. We take the objections to "optimization transfer" well and find...

Quantile Regression via an MM Algorithm.
March 1, 2000... Quantile regression is an increasingly popular method for estimating the quantiles of a distribution conditional on the values of covariates. Regression quantiles are robust against the influence of outliers and, taken several at a time, they...

Fitting Mixed-Effects Models Using Efficient EM-Type Algorithms.
March 1, 2000... In recent years numerous advances in EM methodology have led to algorithms which can be very efficient when compared with both their EM predecessors and other numerical methods (e.g., algorithms based on Newton-Raphson). This article combines...

Importance Link Function Estimation for Markov Chain Monte Carlo Methods.(Statistical Data Included)
March 1, 2000... This article focuses on improving estimation for Markov chain Monte Carlo simulation. The proposed methodology is based upon the use of importance link functions. With the help of appropriate importance sampling weights, effective estimates of...

Adaptive Bayesian Regression Splines in Semiparametric Generalized Linear Models.(Statistical Data Included)
March 1, 2000... This article presents a fully Bayesian approach to regression splines with automatic knot selection in generalized semiparametric models for fundamentally non-Gaussian responses. In a basis function representation of the regression spline we...

Maximum Likelihood for Generalized Linear Models With Nested Random Effects via High-Order, Multivariate Laplace Approximation.(Statistical Data Included)
March 1, 2000... Nested random effects models are often used to represent similar processes occurring in each of many clusters. Suppose that, given cluster-specific random effects b, the data y are distributed according to f(y|b, [Theta]), while b follows a...

Note on "Obtaining the Maximum Likelihood Estimates in Incomplete R x C Contingency Tables Using a Poisson Generalized Linear Model".(Statistical Data Included)
March 1, 2000... This note extends the construction of the design matrix used for estimating cell probabilities with ignorable missing data described by Lipsitz, Parzen, and Molenberghs. A reformulation for the general case of an n-way table is described and...

Two-Sample [T.sub.3] Plot: A Graphical Comparison of Two Distributions.(Statistical Data Included)
March 1, 2000... Consider two independent random variables x and y with means and standard deviations [[Mu].sub.x], [[Mu].sub.y], [[Sigma].sub.x], and [[Sigma].sub.y], respectively. Let [F.sub.x](t) = P[(x - [[Mu].sub.x]) / [[Sigma].sub.x] [is less than or...

Fitting Quantiles: Doubling, HR, HQ, and HHH Distributions.(Statistical Data Included)
March 1, 2000... This article introduces a family of distributional shapes which is flexible in the sense that it contains skewed and symmetric laws as well as heavy-tailed and light-tailed laws. The proposed family is also practically convenient because it is...

Robust Frequency Estimation Using Elemental Sets.(Statistical Data Included)
March 1, 2000... The extraction of sinusoidal signals from time-series data is a classic problem of ongoing interest in the statistics and signal processing literatures. Obtaining least squares estimates is difficult because the sum of squares has local minima...

Correction.(Correction Notice)
March 1, 2000... Yu, Y., and Lambert, D. (1999), "Fitting Trees to Functional Data, With an Application to Time-of-Day Patterns," 8, 749-762; and Grillenzoni, C. (1999), "Adaptive Tests for Changing Unit Roots in Nonstationary Time Series," 8, 763-778. In...

[0] Discussion.
March 1, 2000... We welcome this penetrating discussion of algorithms based on majorization, because we ourselves have found iterative majorization a very convenient minimization method for a broad range of statistical problems (see Heiser 1995). We believe...

[0] Discussion.
March 1, 2000... The general idea of optimization transfer is very appealing to me, especially since I have never succeeded in fully understanding the EM algorithm. I like the examples in Lange, Hunter, and Yang's article and suspect that even further...