DATA-BASED CHOICE OF THE SMOOTHING PARAMETER FOR A KERNEL DENSITY ESTIMATOR

The problem of optimally choosing the smoothing parameter of a Fourier integral density estimator (FIE) is addressed. A new data-based method of so doing is shown to be related to, but an improvement over, the method of Davis (1981). It is shown that Davis' method does not lead to consistent estimation of certain, principally multimodal, densities. In a simulation study involving the bimodal mixture of two normal densities, the new method is seen to represent a substantial improvement.     

SERIES REPRESENTATION OF DOUBLY NONCENTRAL t DISTRIBUTION BY USE OF THE MELLIN INTEGRAL TRANSFORM

ABSTRACT

The Mellin integral transform is widely used to find the distributions of products and quotients of independent random variables defined over the positive domain. But it is hardly used to derive the distributions defined over both positive and negative values of the random variables. In this paper, the Mellin integral transform is applied to obtain the doubly noncentral t density and its distribution function in convergent series forms.     

再生核Hilbert空间展开的函数型回归模型变量选择

针对协变量是函数型、响应变量是标量的多元函数型回归模型,文章提出了函数系数基于再生核Hilbert空间展开的变量选择方法。首先,利用带积分余项的泰勒展开式和再生核Hilbert空间内积性质将模型转化为结构化形式,其次,通过自适应弹性网惩罚对结构化模型中的组间和组内系数同时进行压缩。结果证明了这种压缩估计具有Oracle性质,蒙特卡罗模拟结果也显示新方法在不同样本量、不同噪声和变量相关性干扰下均优于基于普通基函数展开的变量选择方法,且尤其适用于原始协变量高度相关的情形。最后,通过分析一个商品房平均销售价格影响因素数据演示了新方法的应用。     

A unified approach to minimum variance unbiased estimation of a probability function belonging to an exponential family

The problem of minimum variance unbiased estimation of the probability density function of a random variable belonging to an exponential family is considered. The method of estimation proposed in this paper requires the solution of a certain integral equation. For many probability distributions the solution of this equation is given by a known result in integral transform theory.     

Computing the normalization factor in bingham distributions

A new method for calculating the confluent hypergeometric function of vector argument is presented. This function serves to normalize the integral which defines the Bingham distribution. By using a polynomial and series expansion, the double integral can be reduced to a single infinite summation whose coefficients may be computed recursively. A short c program implementing the method is included in the appendix.     

Calibrated path sampling and stepwise bridge sampling

A computational problem in many fields is to evaluate multiple integrals and expectations simultaneously. Consider probability distributions with unnormalized density functions indexed by parameters on a 2-dimensional grid, and assume that samples are simulated from distributions on a subgrid. Examples of such unnormalized density functions include the observed-data likelihoods in the presence of missing data and the prior times the likelihood in Bayesian inference. There are various methods using a single sample only or multiple samples jointly to compute each integral. Path sampling seems a compromise, using samples along a 1-dimensional path to compute each integral. However, different choices of the path lead to different estimators, which should ideally be identical. We propose calibrated estimators by the method of control variates to exploit such constraints for variance reduction. We also propose biquadratic interpolation to approximate integrals with parameters outside the subgrid, consistently with the calibrated estimators on the subgrid. These methods can be extended to compute differences of expectations through an auxiliary identity for path sampling. Furthermore, we develop stepwise bridge-sampling methods in parallel but complementary to path sampling. In three simulation studies, the proposed methods lead to substantially reduced mean squared errors compared with existing methods.     

The computation of standard normal distribution integral in any required precision based on reliability method

Based on reliability theory, the value of the standard normal distribution integral can be obtained by calculating the probability of the failure domain of the linear performance function. After the sample space is divided into some sub-sample spaces, a number of sub-failure domains are obtained. In the paper, the methods of computing the probabilities of sub-failure domains are discussed. All the formulae and the steps of computing the standard normal distribution integral which meet any required precision are given in the paper. Examples show that it is easy for the method to compute the standard normal distribution integral.     

Using the Fast Fourier Transform to Compute Multiple Comparisons With the Best and Subset Selection Critical Values

Multiple comparison methods are widely implemented in statistical packages and heavily used. To obtain the critical value of a multiple comparison method for a given confidence level, a double integral equation must be solved. Current computer implementations evaluate one double integral for each candidate critical value using Gaussian quadrature. Consequently, iterative refinement of the critical value can slow the response time enough to hamper interactive data analysis. However, for balanced designs, to obtain the critical value for multiple comparisons with the best, subset selection, and one-sided multiple comparison with a control, if one regards the inner integral as a function of the outer integration variable, then this function can be obtained by discrete convolution using the Fast Fourier Transform (FFT). Exploiting the fact that this function need not be re-evaluated during iterative refinement of the critical value, it is shown that the FFT method obtains critical values at least four times as accurate and two to five times as fast as the Gaussian quadrature method.     

Importance sampling for Kolmogorov backward equations

The solution of the Kolmogorov backward equation is expressed as a functional integral by means of the Feynman–Kac formula. The expectation value is approximated as a mean over trajectories. In order to reduce the variance of the estimate, importance sampling is utilized. From the optimal importance density, a modified drift function is derived which is used to simulate optimal trajectories from an Itô equation. The method is applied to option pricing and the simulation of transition densities and likelihoods for diffusion processes. The results are compared to known exact solutions and results obtained by numerical integration of the path integral using Euler transition kernels. The importance sampling leads to strong variance reduction, even if the unknown solution appearing in the drift is replaced by known reference solutions. In models with low-dimensional state space, the numerical integration method is more efficient, but in higher dimensions it soon becomes infeasible, whereas the Monte Carlo method still works.     

Generating correlated random vector involving discrete variables

For the issue of generating correlated random vector containing discrete variables, one major obstacle is to determine a suitable correlation coefficient ρ_z in normal space for a specified correlation coefficient ρ_x. This paper develops a method to solve this problem. First, the double integral evaluated for ρ_x is transformed into independent standard uniform space, then, a Quasi Monte Carlo method is introduced to calculate the double integral. For a given ρ_x, an appropriate ρ_z is determined by a false position method. Compared with existing methodologies, the proposed method is less efficient, but it is relatively easy to implement.     

A CLOSED-FORM EXPRESSION FOR THE PEARSON TYPE IV DISTRIBUTION FUNCTION

A closed-form expression is presented for the probability integral of the Pearson Type IV distribution, and a corresponding method of evaluation is given. This analysis addresses a long-standing gap in the theory of the Pearson system of distributions. In addition, a simple derivation is given of an expression for the normalizing constant in the Type IV integral.     

THE MONTE CARLO EVALUATION OF ORTHANT PROBABILITIES FOR MULTIVARIATE NORMAL DISTRIBUTIONS

The calculation of multivariate normal orthant probabilities is practically impossible when the number of variates is greater than five or six, except in very special cases. A transformation of the integral is obtained which enables quite accurate Monte Carlo estimates to be obtained for a fairly high number of dimensions, particularly if control variates are used.     

A Monte Carlo method for filtering a marked doubly stochastic Poisson process

The author provides an approximated solution for the filtering of a state-space model, where the hidden state process is a continuous-time pure jump Markov process and the observations come from marked point processes. Each state k corresponds to a different marked point process, defined by its conditional intensity function λ _k (t). When a state is visited by the hidden process, the corresponding marked point process is observed. The filtering equations are obtained by applying the innovation method and the integral representation theorem of a point process martingale. Since the filtering equations belong to the family of Kushner–Stratonovich equations, an iterative solution is calculated. The theoretical solution is approximated and a Monte Carlo integration technique employed to implement it. The sequential method has been tested on a simulated data set based on marked point processes widely used in the statistical analysis of seismic sequences: the Poisson model, the stress release model and the Etas model.     

Likelihood‐based Inference with Missing Data Under Missing‐at‐Random

Likelihood‐based inference with missing data is challenging because the observed log likelihood is often an (intractable) integration over the missing data distribution, which also depends on the unknown parameter. Approximating the integral by Monte Carlo sampling does not necessarily lead to a valid likelihood over the entire parameter space because the Monte Carlo samples are generated from a distribution with a fixed parameter value. We consider approximating the observed log likelihood based on importance sampling. In the proposed method, the dependency of the integral on the parameter is properly reflected through fractional weights. We discuss constructing a confidence interval using the profile likelihood ratio test. A Newton–Raphson algorithm is employed to find the interval end points. Two limited simulation studies show the advantage of the Wilks inference over the Wald inference in terms of power, parameter space conformity and computational efficiency. A real data example on salamander mating shows that our method also works well with high‐dimensional missing data.     

ON DERIVATION OF INITIAL BLOCKS OF B.I.B. DESIGNS WITH MORE THAN ONE INITIAL BLOCK

A good amount of work has been done on the construction of balanced incomplete block (B.I.B.) designs by Bose (1939, 1942), Sprott (1954, 1956), Rao (1961), Takeuchi (1962) and others. Sprott (1954, 1956) obtained several series of B.I.B. designs through difference sets. The main purpose of the present investigation is to provide two methods of construction of B.I.B. designs obtainable through more than one initial block. The first method derives initial blocks of a series of designs from some specified blocks of a B.I.B. design obtainable by developing one or more initial blocks. The second method attempts to obtain one of the initial blocks (the basic initial block) through the different powers of an element of a finite field; then an appropriate method for generating the other initial blocks from it is discussed. A table showing the basic initial block for different designs has been presented. By these methods several solutions of some B.I.B. designs could be obtained from different initial blocks. An examination was therefore made to see if these designs were all isomorphic.     

The lognormal characteristic function

A number of different ways are examined of representing the characteristic function φ(t) of the lognormal distribution, which cannot be expanded in a Taylor series based on the moments. In §2 the use of a finite Taylor series is examined. A method of summing the divergent formal expansion is discussed in §3. In §4 the fact that φ(t) is a boundary analytic function is exploited. Asymptotic approximation of the integral defining φ(t) is studied in §5. Each approach produces some interesting information about the distribution.     

Exact simulation of tempered stable Ornstein–Uhlenbeck processes

Based on a representation of a stochastic integral of Ornstein–Uhlenbeck (O–U) type, the exact simulation algorithm of the tempered stable O–U process is given in this paper. The algorithm employs the double rejection method and the general acceptance–rejection technique. The time complexity of the double rejection method is uniformly bounded over all values of the parameter. And the acceptance probability of the acceptance–rejection technique can be improved to as close to 1 as possible. Thus, the implementation of the algorithm is efficient. The performance of the simulation method is evidenced by some empirical results.     

GIL-PELAEZ-ROSÉN TRANSFORMATION OF DATA

Several statistics based on the empirical characteristic function have been proposed for testing the simple goodness-of-fit hypothesis that the data come from a population with a completely specified characteristic function which cannot be inverted in a closed form, the typical example being the class of stable characteristic functions. As an alternative approach, it is pointed out here that the inversion formula of Gil-Pelaez and Rosén, as applied to the data and the hypothetical characteristic function via numerical integration, is the natural replacement of the probability integral transformation in the given situation. The transformed sample is from the uniform (0, l) distribution if and only if the null hypothesis is true, and for testing uniformity on (0,1) the whole arsenal of methods statistics so far produced can be used.     

ON THE PERFORMANCE OF KOLMOGOROV-SMIRNOV TESTS ON STABLE RANDOM VARIABLES1

This paper provides two procedures to perform the Kolmogorov–Smirnov (K-S) tests on stable random variables. One utilizes the integral representation of the cumulative distribution function of this random variable to perform the K-S test; the other utilizes the Gil-Pelaez-Roseń transformation of data (Csörg(1983)) to test uniformity on (0,1) for the transformed sample. Both procedures are examined and evaluated by a simulation study.     

Volume estimation by monte carlo methods *

In estimating a multiple integral, it is known that Monte Carlo methods are more efficient than analytical techniques when the number of dimensions is beyond seven. In general, the sample-mean method is better than the hit-or-miss Monte Carlo method. However, when the volume of a domain in a high-dimensional space is of interest, the hit-or-miss method is usually preferred. It is because of the difficulty in generalizing the sample-mean method for the computation of the volume of a domain. This paper develops a technique to make such a generalization possible. The technique can be interpreted as a volume-preserving transformation procedure. A volume-preserving transformation is first performed to transform the concerned domain into a hypersphere. The volume of the domain is then evaluated by computing the volume of the hypersphere.