In spite of increasingly sophisticated workforce management tools, a significant gap remains between the goal of effective staffing and the present difficulty predicting the stochastic demand of inbound calls. We have investigated the hypothesized nonhomogeneous Poisson process model of modem pool callers of the University community. In our case, we tested if the arrivals could be approximated by a piecewise constant rate over short intervals. For each of 1 and 10-minute intervals, based on the close relationship between the Poisson process and the exponential distribution, the test results did not show any sign of homogeneous Poisson process. We have examined the hypothesis of a nonhomogeneous Poisson process by a transformed statistic. Quantitative and graphical goodness-of-fit tests have confirmed nonhomogeneous Poisson process. <p>Further analysis on the intensity function revealed that linear rate intensity was woefully inadequate in predicting time varying arrivals. For sinusoidal rate model, difficulty arose in setting the period parameter. Spline models, as an alternative to parametric modelling, had more control of balance between data fitting and smoothness, which was appealing to our analysis on call arrival process.

In spite of increasingly sophisticated workforce management tools, a significant gap remains between the goal of effective staffing and the present difficulty predicting the stochastic demand of inbound calls. We have investigated the hypothesized nonhomogeneous Poisson process model of modem pool callers of the University community. In our case, we tested if the arrivals could be approximated by a piecewise constant rate over short intervals. For each of 1 and 10-minute intervals, based on the close relationship between the Poisson process and the exponential distribution, the test results did not show any sign of homogeneous Poisson process. We have examined the hypothesis of a nonhomogeneous Poisson process by a transformed statistic. Quantitative and graphical goodness-of-fit tests have confirmed nonhomogeneous Poisson process. <p>Further analysis on the intensity function revealed that linear rate intensity was woefully inadequate in predicting time varying arrivals. For sinusoidal rate model, difficulty arose in setting the period parameter. Spline models, as an alternative to parametric modelling, had more control of balance between data fitting and smoothness, which was appealing to our analysis on call arrival process.

Indiana University-Purdue University Indianapolis (IUPUI) / A semi-parametric approach incorporates parametric and nonparametric functions in the model and is very useful in situations when a fully parametric model is inadequate. The objective of this dissertation is to extend statistical methodology employing the semi-parametric modeling approach to analyze data in health science research areas. This dissertation has three parts. The first part discusses the modeling of the dose-response relationship with correlated data by introducing overall drug effects in addition to the deviation of each subject-specific curve from the population average. Here, a penalized spline regression method that allows modeling of the smooth dose-response relationship is applied to data in studies monitoring malaria drug resistance through the ex-vivo assays.The second part of the dissertation extends the SiZer map, which is an exploratory and a powerful visualization tool, to detect underlying significant features (increase, decrease, or no change) of the curve at various smoothing levels. Here, Penalized Spline Significant Zero Crossings of Derivatives (PS-SiZer), using a penalized spline regression, is introduced to investigate significant features in correlated data arising from longitudinal settings. The third part of the dissertation applies the proposed PS-SiZer methodology to analyze HIV data. The durability of significant weight change over a period is explored from the PS-SiZer visualization. PS-SiZer is a graphical tool for exploring structures in curves by mapping areas where rate of change is significantly increasing, decreasing, or does not change. PS-SiZer maps provide information about the significant rate of weigh change that occurs in two ART regimens at various level of smoothing. A penalized spline regression model at an optimum smoothing level is applied to obtain an estimated first-time point where weight no longer increases for different treatment regimens.

Optimization of simulation model performance requires finding the values of the model's controllable inputs that optimize a chosen model response. Responses are usually stochastic in nature, and the cost of simulation model runs is high. The literature suggests the use of metamodels to synthesize the response surface using sample data. In particular, nonparametric regression is proposed as a useful tool in the global optimization of a response surface. As the general simulation optimization problem is very difficult and requires expertise from a number of fields, there is a growing consensus in the literature that a knowledge-based approach to solving simulation optimization problems is required. This dissertation examines the relative performance of the principal nonparametric techniques, spline and kernel smoothing, and subsequently addresses the issues involved in implementing the techniques in a knowledge-based simulation optimization system. The dissertation consists of two parts. In the first part, a full factorial experiment is carried out to compare the performance of kernel and spline smoothing on a number of measures when modeling a varied set of surfaces using a range of small sample sizes. In the second part, nonparametric metamodeling techniques are placed in a taxonomy of stochastic search procedures for simulation optimization and a method for their implementation in a knowledge-based system is presented. A sequential design procedure is developed that allows spline smoothing to be used as a search technique. Throughout the dissertation, a two-input, single-response model is considered. Results from the experiment show that spline smoothing is superior to constant-bandwidth kernel smoothing in fitting the response. Kernel smoothing is shown to be more accurate in placing optima in X-space for sample sizes up to 36. Inventory model examples are used to illustrate the results. The taxonomy implies that search procedures can be chosen initially using the parameters of the problem. A process that allows for selection of a search technique and its subsequent evaluation for further use or for substitution of another search technique is given. The success of a sequential design method for spline smooths in finding a global optimum is demonstrated using a bimodal response surface. / Ph. D.

Neste trabalho, apresentamos os modelos de regressão spline de nós-livres como uma alternativa aos modelos não lineares utilizados em curvas de crescimento multifásico. Estudaremos o algoritmo de busca cega a través da seção dourada para escolher a melhor quantidade de nós e suas respectivas localidades. O pacote freeknotspline do software livre R foi utilizado para ajustar os modelos propostos. O critério de informação de Akaike foi usado para escolher o melhor modelo para diferentes graus do polinômio associados ao spline. Estudos de simulação foram realizados para entender melhor a posição dos nós, localidade e grau do polinômio relacionado aos modelos de regressão spline de nós-livres e como isto pode afetar a qualidade de ajuste do modelo. Com base no nosso estudo de simulação, propomos uma forma empírica de determinar o numero de nós, deixando que o algoritmo de busca escolha a posição dos nós. A metodologia é aplicada aos dados de crescimento multifásico de vacas da raça Hereford. / In this work, we present the free-knot spline regression models as an alternative to the nonlinear models used in multiphase growth curves. We will study the blind search algorithm through the gold section to choose the best number of knot and their respective locations. The package freeknotspline of the free software R was used to fit the proposed models. The Akaike information criterion was used to choose the best model for different degrees of the polynomial associated with the spline. Simulation studies were performed to better understand the position of the knot, location and degree of the polynomial related to spline regression models of free-knots, and how this may affect the goodness of fit of the model. Based on our simulation study, we propose an empirical way of determining the number of knots, letting the search algorithm choose the position of the knots. The methodology is applied to the multiphase growth data of Hereford breed females.

In this thesis, Bayesian semiparametric models for the missing data mechanisms of nonignorably missing covariates in logistic regression are developed. In the missing data literature, fully parametric approach is used to model the nonignorable missing data mechanisms. In that approach, a probit or a logit link of the conditional probability of the covariate being missing is modeled as a linear combination of all variables including the missing covariate itself. However, nonignorably missing covariates may not be linearly related with the probit (or logit) of this conditional probability. In our study, the relationship between the probit of the probability of the covariate being missing and the missing covariate itself is modeled by using a penalized spline regression based semiparametric approach. An efficient Markov chain Monte Carlo (MCMC) sampling algorithm to estimate the parameters is established. A WinBUGS code is constructed to sample from the full conditional posterior distributions of the parameters by using Gibbs sampling. Monte Carlo simulation experiments under different true missing data mechanisms are applied to compare the bias and efficiency properties of the resulting estimators with the ones from the fully parametric approach. These simulations show that estimators for logistic regression using semiparametric missing data models maintain better bias and efficiency properties than the ones using fully parametric missing data models when the true relationship between the missingness and the missing covariate has a nonlinear form. They are comparable when this relationship has a linear form.

The aim of this thesis is to provide an overview of the varying coefficient mod- els - a class of regression models that allow the coefficients to vary as functions of random variables. This concept is described for independent samples, longi- tudinal data, and time series. Estimation methods include polynomial spline, smoothing spline, and local polynomial methods for models of a linear form and local maximum likelihood method for models of a generalized linear form. The statistical properties focus on the consistency and asymptotical distribution of the estimators. The numerical study compares the finite sample performance of the estimators of coefficient functions. 1

本研究探討常用的兩個無母數迴歸方法，核迴歸與樣條迴歸，在具有傘型限制式下，對於傘型函數的估計與不具限制式下的傘型函數估計比較，同時也探討不同誤差變異對估計結果的影響，並進一步探討受限制下兩方法的估計比較。本研究採用「估計頂點位置與實際頂點位置差」及「誤差平方和」作為衡量估計結果的指標。在帶寬及節點的選取上，本研究採用逐一剔除交互驗證法來篩選。模擬結果顯示，受限制的核函數在誤差變異較大的頂點位置估計較佳，誤差變異縮小時反而頂點位置估計較差，受限制的B-樣條函數也有類似的狀況。而在兩方法的比較上，對於較小的誤差變異，核函數的頂點位置估計能力不如樣條函數，但在整體的誤差平方和上卻沒有太大劣勢，當誤差變異較大時，核函數的頂點位置估計能力有所提升，整體誤差平方和仍舊維持還不錯的結果。 / In this study, we give an umbrella order constraint on kernel and spline regression model. We compare their estimation in two measurements, one is the difference of estimate peak and true peak, the other one is the sum of square difference on predict and the true value. We use leave-one-out cross validation to select bandwidth for kernel function and also to decide the number of knots for spline function. The effect of different error size is also considered. Some of R packages are used when doing simulation. The result shows that when the error size is bigger, the prediction of peak location is better in both constrained kernel and spline estimation. The constrained spline regression tends to provide better peak location estimation compared to constrained kernel regression.