Nonparametric estimation of a change in defect intensity

Summary.  In an important class of problems involving mixture distributions, interest focuses on the mixture proportions, considering other possible parameters as nuisance parameters. We formulate a new variation on such problems that arose in a study on the link between the number of cells in a charge-coupled detector image sensor that turned defective because of cosmic radiation and the storage conditions of such sensors. In this variation, the component densities are bivariate, there are two classes and only a subset of the mixture proportions is of relevance. We propose a nonparametric method to estimate the relevant subset of the proportions and apply our method to the data in our study.     

技术工人满意度影响因素的实证分析

技术工人的短缺是影响中国经济以更快速度发展的障碍之一。而难以吸引更多人才投身技术岗位的重要原因之一是对技术工人的激励不足,造成其满意度较低。文章以问卷调查为基础,利用多元方差分析法对技术工人满意度的影响因素进行了实证研究,并根据研究结果给出一些建议,希望为企业根据不同技术工人群体制定适宜的激励政策提供参考。     

Semiparametric mixture models and repeated measures: the multinomial cut point model

Summary.  Suppose that we have m repeated measures on each subject, and we model the observation vectors with a finite mixture model.  We further assume that the repeated measures are conditionally independent. We present methods to estimate the shape of the component distributions along with various features of the component distributions such as the medians, means and variances. We make no distributional assumptions on the components; indeed, we allow different shapes for different components.     

The predictability of stock returns - a nonparametric approach

This paper reexamines the predictability of stock returns with a nonparametric model. We first identify, through a set of diagnostic tests, five lagged predictive factors from a linear model. Using these factors, we predict one-month-ahead stock index returns with a nonparametric approach. We find that our nonparametricmodel. We first identify, through a set of diagnostic tests, five lagged predictive factors from a linear model. Using these factors, we predict on -month-ahead stock index returns with a nonparametric approach. We find that our nonparametric model can correctly predict about 74% of stock index return signs. With various ex ante trading rules based on nonparametric predictions and transaction cost schedules, we then compare the performance of "managed" portfolios with that of the buy and hold portfolios. We fmd that the managed portfolios are mean-variance dominant over the buy-and-hold strategies when no or low transaction costs are assumed. When high transaction costs are assumed instead, the mean-variance dominance diminishes However,the Sharpe index of risk-adjusted portfolio performanceindicates that the managed portfolios significantly outperform the buy-and-hold strategies even for the high-transaction cost scenario. We show that the difference in performance between the managed portfolios and the buy-and-hold strategies can be partially explained by the January effect or the small firm effect. In sum, this paper demonstrates the merits of using a nonparametric approach for predicting stock returns and testing market efficiency.     

Estimators of shift based on statistics of the Kolmogorov-Smirnov type

This paper is concerned with the estimation of a shift parameter δ_o, based on some nonnegative functional Hg₁ of the pair (D^δ_N(x), f?^δ_N(x)), where D^δ_N(x) = K_N/b {F_2,n(x)—F_1,m (x + δ)}, +^δ_N(x) = {mF_1,m (x + δ) + nF_2,n(x)}/N, where F_1,m and F_2,n are the empirical distribution functions of two independent random samples (N = m + n), and where K²_N = mn/N. First an estimator δ_N, is defined as a value of δ minimizing a functional H of the type of H₁. A second estimator δ¹_N is also defined which is a linearized version of the first. Finite and asymptotic properties of these estimators are considered. It is also shown that most well-known test statistics of the Kolmogorov-Smirnov type are particular cases of such functionals H₁. The asymptotic distribution and the asymptotic efficiency of some estimators are given.     

Nonparametric curve estimation with missing data: A general empirical process approach

A general nonparametric imputation procedure, based on kernel regression, is proposed to estimate points as well as set- and function-indexed parameters when the data are missing at random (MAR). The proposed method works by imputing a specific function of a missing value (and not the missing value itself), where the form of this specific function is dictated by the parameter of interest. Both single and multiple imputations are considered. The associated empirical processes provide the right tool to study the uniform convergence properties of the resulting estimators. Our estimators include, as special cases, the imputation estimator of the mean, the estimator of the distribution function proposed by Cheng and Chu [1996. Kernel estimation of distribution functions and quantiles with missing data. Statist. Sinica 6, 63–78], imputation estimators of a marginal density, and imputation estimators of regression functions.     

The probability of degenerate maximum-likelihood estimates with randomly truncated survival data

We study the nonparametric maximum-likelihood estimator of the marginal distribution function in bivariate data with truncated sum (Woodroofe 1985), and calculate the probability that it will assign zero probability to a range of observed values. We show that in a simple artificial model with uniform intensity in both variables the probability may be expressed using recursive formulae, and that large-sample approximations valid under weak regularity conditions show good agreement with as few as 20 observations. In the uniform-intensity model, the probability of degenerate estimates is rather small, being just over 5% with a dataset of 20 observations and falling below 1% with 125 observations. However, in a model of transfusion-associated AIDS in 125 children and adults (Kalbfleisch and Lawless 1989), the probability of degenerate estimates is much larger—about 30% with 20 observations and 3% with the actual 125 observations. This is largely because both the infection and incubation density are increasing over time. Such a pattern makes the condition for degeneracy, i.e., relatively high observed intensity at extreme values of each variable, much more likely.     

A New Method for Testing Interaction in Unreplicated Two-Way Analysis of Variance

For a two-way ANOVA table, with a single observation per cell, the standard approach is to assume that interaction between the two factors is negligible, and to base inferences about the main factors on the model without interaction. But there is no totally satisfactory method for testing if interaction can be ignored. The classical approach is to specify a functional form for the interaction terms, involving a small number of parameters, and then use an appropriate test. But, such tests have low power if the functional form is inappropriate. This has led researchers to propose tests which do not assume a specific form for the interactions. In this article, we present a new approach for testing interaction which also does not assume a specific form for the interaction. This approach is fairly simple and flexible, and its usefulness is illustrated with several examples. We also present a general result which shows that there is no test of interaction with good power properties against all types of interaction.     

Testing for the Equality of Two Autoregressive Functions Using Quasi-Residuals

This article discusses the problem of testing the equality of two nonparametric autoregressive functions against one-sided alternatives. The heteroscedastic errors and stationary densities of the two independent strong mixing strictly stationary time series can be possibly different. The article adapts the idea of using sum of quasi-residuals to construct the test and derives its asymptotic null distributions. The article also shows that the test is consistent for general alternatives and obtains its limiting distributions under a sequence of local alternatives. Then a Monte Carlo simulation is conducted to study the finite sample level and power behavior of these tests at some alternatives. We also compare the test to an existing lag matched test theoretically and by Monte Carlo experiments.