Structure discovery and parametrically guided regression

In this article, we consider semiparametric estimation in a partially linear single-index panel data model with fixed effects. Without taking the difference explicitly, we propose using a semiparametric minimum average variance estimation (SMAVE) based on a dummy variable method to remove the fixed effects and obtain consistent estimators for both the parameters and the unknown link function. As both the cross-section size and the time series length tend to infinity, we not only establish an asymptotically normal distribution for the estimators of the parameters in the single index and the linear component of the model, but also obtain an asymptotically normal distribution for the nonparametric local linear estimator of the unknown link function. The asymptotically normal distributions of the proposed estimators are similar to those obtained in the random effects case. In addition, we study several partially linear single-index dynamic panel data models. The methods and results are augmented by simulation studies and illustrated by application to two real data examples. This article has online supplementary materials.

The systematic risk estimation models: A different perspective

Recent works have shown that the resting-state brain functional connectivity hypernetwork, where multiple nodes can be connected, are an effective technique for brain disease diagnosis and classification research. The lasso method was used to construct hypernetworks by solving sparse linear regression models in previous research. But, constructing a hypernetwork based on the lasso method simply selects a single variable, in that it lacks the ability to interpret the grouping effect. Considering the group structure problem, the previous study proposed to create a hypernetwork based on the elastic net and the group lasso methods, and the results showed that the former method had the best classification performance. However, the highly correlated variables selected by the elastic net method were not necessarily in the active set in the group. Therefore, we extended our research to address this issue. Herein, we propose a new method that introduces the sparse group lasso method to improve the construction of the hypernetwork by solving the group structure problem of the brain regions. We used the traditional lasso, group lasso method, and sparse group lasso method to construct a hypernetwork in patients with depression and normal subjects. Meanwhile, other clustering coefficients (clustering coefficients based on pairs of nodes) were also introduced to extract features with traditional clustering coefficients. Two types of features with significant differences obtained after feature selection were subjected to multi-kernel learning for feature fusion and classification using each method, respectively. The network topology results revealed differences among the three networks, where hypernetwork using the lasso method was the strictest; the group lasso, most lenient; and the sgLasso method, moderate. The network topology of the sparse group lasso method was similar to that of the group lasso method but different from the lasso method. The classification results show that the sparse group lasso method achieves the best classification accuracy by using multi-kernel learning, which indicates that better classification performance can be achieved when the group structure exists and is properly extended.

There is a long-standing debate in the statistical, epidemiological, and econometric fields as to whether nonparametric estimation that uses machine learning in model fitting confers any meaningful advantage over simpler, parametric approaches in finite sample estimation of causal effects. We address the question: when estimating the effect of a treatment on an outcome, how much does the choice of nonparametric vs parametric estimation matter? Instead of answering this question with simulations that reflect a few chosen data scenarios, we propose a novel approach to compare estimators across a large number of data-generating mechanisms drawn from nonparametric models with semi-informative priors. We apply this proposed approach and compare the performance of two nonparametric estimators (Bayesian adaptive regression tree and a targeted minimum loss-based estimator) to two parametric estimators (a logistic regression-based plug-in estimator and a propensity score estimator) in terms of estimating the average treatment effect across thousands of data-generating mechanisms. We summarize performance in terms of bias, confidence interval coverage, and mean squared error. We find that the two nonparametric estimators can substantially reduce bias as compared to the two parametric estimators in large-sample settings characterized by interactions and nonlinearities while compromising very little in terms of performance even in simple, small-sample settings.

In this article, a semiparametric time‐varying nonlinear vector autoregressive (NVAR) model is proposed to model nonlinear vector time series data. We consider a combination of parametric and nonparametric estimation approaches to estimate the NVAR function for both independent and dependent errors. We use the multivariate Taylor series expansion of the link function up to the second order which has a parametric framework as a representation of the nonlinear vector regression function. After the unknown parameters are estimated by the maximum likelihood estimation procedure, the obtained NVAR function is adjusted by a nonparametric diagonal matrix, where the proposed adjusted matrix is estimated by the nonparametric kernel estimator. The asymptotic consistency properties of the proposed estimators are established. Simulation studies are conducted to evaluate the performance of the proposed semiparametric method. A real data example on short‐run interest rates and long‐run interest rates of United States Treasury securities is analyzed to demonstrate the application of the proposed approach.The Canadian Journal of Statistics47: 668–687; 2019 © 2019 Statistical Society of Canada

In this article, we introduce two new test statistics based on pairwise distance to check whether a parametric model with missing response is correctly specified or not. The proposed test statistics are easy to implement and no non parametric function estimation is involved. The asymptotic properties of the newly introduced test statistics are investigated. Under the null hypothesis, it is shown that they converge to a non degenerate distribution. And it can also detect any local alternatives that are different from the null at the parametric rate. In addition, the convergence rates of the proposed test are unaffected by the dimension of covariates. Some simulation studies and a real data example are conducted to show the good performance of the proposed test.

In the problem of estimating an unknown distribution function $F$ in the presence of censoring, one can use a nonparametric estimator such as the Kaplan-Meier estimator, or one can consider parametric modeling. There are many situations where physical reasons indicate that a certain parametric model holds approximately. In these cases a nonparametric estimator may be very inefficient relative to a parametric estimator. On the other hand, if the parametric model is only a crude approximation to the actual model, then the parametric estimator may perform poorly relative to the nonparametric estimator, and may even be inconsistent. The Bayesian paradigm provides a reasonable framework for this problem. In a Bayesian approach, one would try to put a prior distribution on $F$ that gives most of its mass to small neighborhoods of the entire parametric family. We show that certain priors based on the Dirichlet process prior can be used to accomplish this. For these priors the posterior distribution of $F$ given the censored data appears to be analytically intractable. We provide a method for approximating this posterior distribution through the use of a successive substitution sampling algorithm. We also show convergence of the algorithm.

A control chart is one of the statistical process techniques that is used to monitor different processes. Some processes are characterized by functions or profiles, and a profile is a functional relationship between the dependent and independent variable(s) used to monitor the quality of the process. Several research studies were conducted on linear profiling where only fixed effects are considered. However, in this research, we focus on random effects as they represent the differences between profiles and thus are more proper for interpretation. Two approaches are proposed in this study for Phase II profile monitoring; the first approach is the nonparametric via residuals and the second is the semiparametric approach, where this technique combines the parametric estimates with a portion of the nonparametric estimates to the residuals. Usually, parametric estimations lead to biased estimates when the model is misspecified, whereas nonparametric estimates may give high variances, and thus semiparametric estimates are preferred. New nonparametric and semiparametric multivariate exponential weighted moving average (MEWMA) control charts are introduced and their performances compared to the parametric approach for different samples and shift sizes, and the correlation between and within profiles was considered. The average run length (ARL) and average time to signal (ATS) criteria are used for choosing the best approach. Simulation studies and real datasets were utilized for comparing the performance of the proposed MEWMA charts.

ملخص:- تحتوي نماذج الانحدار على مکونين أحدهما مکون عشوائي يمثل حد الخطأ (Error Term) والأخر مکون غير عشوائي يمثل متغير الاستجابة(Response Variable) وهذا المکون هو محل الاهتمام فيتم تقديره بإحدى الطريقتين هما: الطريقة المعلمية أو الطريقة اللامعلمية أو کلاهما. نموذج الانحدار المعلمي (Parametric Regression Model) هو النموذج الأکثر شيوعا حيث أنه يفترض أن متغير الاستجابة له صيغه دالية محدده مسبقا والتي تم الحصول عليها من معلومات سابقه حول هيکل دالة الانحدار من خلال ذلک يتم تقدير فئة من المعالم. ورغم أن النماذج المعلمية تعد أحد أهم أدوات تحليل البيانات؛ لأنها فعاله ويمکن تفسيرها بسهوله، إلا انها لم تستطع ان تکون کافية في بعض الحالات. أما نموذج الانحدار اللامعلمي (Nonparametric Regression Model) فانه يعتمد على تقدير دالة الانحدار مباشرة من البيانات کما أنه لا يفترض فروضا معينه لتقدير معالم النموذج مثل التي يفترضها النموذج المعلمي وبالتالي فهو يتمتع بمرونة أکثر لاکتشاف البيانات التي قد تکون مفقودة او في الحالات التي لا تتوافر فيها معلومات سابقة. يتم انشاء نموذج الانحدار شبه المعلمي (Semi parametric Regression Model) من خلال مزج الطريقتين السابقتين معا. فغالبا ما يتم استخدام هذا النموذج في المواقف التي تکون فيها الافتراضات المعلمية غير محددة وغير متسقة، أو أن النموذج اللامعلمي لا يعمل بشکل کامل. يشتمل هذا البحث على تطبيق نماذج الانحدار المعلمية وشبه المعلمية بالإضافة إلى النموذج اللامعلمى کحالة خاصة من نموذج الانحدار شبه المعلمى لتقدير الدالة الافتراضية وتحديد شکل العلاقة بين المتغير التابع والمتغيرات المستقلة بالاضافة الى شکل انتشار البيانات مع خط (أو منحني) التقدير وذلک على بيانات واقعية تم الحصول عليها من الإتحاد المصري لألعاب القوى والمعروفة بـبيانات لعبة العشاري . Regression models contain two components, one of which is a random component that represents the Error Term and the other is a non-random component that represents the response variable and this component is of interest and is estimated in one of two methods: the parametric method or the nonparametric method or both. Parametric regression model is the most common model since it assumes that the response variable has a predetermined functional formula which was obtained from previous information about the structure of the regression function by which a class of parameters is estimated. Although parameter models are one of the most important tools for analyzing data; Because it is effective and can be explained easily, but it could not be sufficient in some cases. As for the nonparametric regression model, it relies on estimating the regression function directly from the data, as it does not assume specific assumptions for estimating the model parameters, such as those assumed by the parametric model, and thus it has more flexibility to discover data that may be missing or in cases where previous information is not available. The semi-parametric regression model is constructed by blending the two previous methods together. This model is often used in situations where the parametric assumptions are indeterminate and inconsistent, or the nonparametric model is not fully functional. This research includes the application of parametric and semi-parametric regression models in addition to the nonparametric model as a special case of the semi-parametric regression model for estimating the hypothetical function and determining the form of the relationship between the dependent variable and the independent variables in addition to the data spread form with a line (or curve) estimation on realistic data On it from the Egyptian Athletics Federation, known as Decathlon data.

Time series clustering is to assign a set of time series into groups that share certain similarity. It has become an attractive analytic tool as many applications require such classifications. Clustering may also result in more accurate parameter estimates when a group of time series are assumed to share common models and parameters, especially for short panel time series. Many existing time series clustering methods are based on the assumption that the time series are linear. However, linearity assumptions often fail to hold. In this paper we consider the problem of clustering nonlinear time series. We propose the use of a two dimensional Kolmogorov-Smirnov statistic as a distance measure of two time series by measuring the affinity of nonlinear serial dependence structures. It is nonparametric in nature hence no model assumption are needed. The approach is illustrated with simulation studies as well as real data examples.

Semiparametric estimation for count data through weighted distributions

Non-invasive recordings of human brain activity through electroencephalography (EEG) or magnetoencelphalography (MEG) are of value for both basic science and clinical applications in sensory, cognitive, and affective neuroscience. Here we introduce a new approach to estimating the intra-cranial sources of EEG/MEG activity measured from extra-cranial sensors. The approach is based on the group lasso, a sparse-prior inverse that has been adapted to take advantage of functionally-defined regions of interest for the definition of physiologically meaningful groups within a functionally-based common space. Detailed simulations using realistic source-geometries and data from a human Visual Evoked Potential experiment demonstrate that the group-lasso method has improved performance over traditional ℓ2 minimum-norm methods. In addition, we show that pooling source estimates across subjects over functionally defined regions of interest results in improvements in the accuracy of source estimates for both the group-lasso and minimum-norm approaches.

In maximum likelihood estimation, the real but unknown innovation distribution is often replaced by a nonparametric estimate, and thus the estimation procedure becomes semiparametric. These semiparametric approaches generally involve two steps: the first step that incorporate an initial estimate of the model parameter to produce a residual sample, and the second step that uses the residuals to estimate the likelihood, which is subsequently maximized to obtain the final estimate of the model parameter. Therefore, the characteristics of the initial input estimator may be carried over to the final semiparametric estimator, and the performance of the semiparametric estimator will be impaired if the input estimate is deficient. In this article we have studied a onestep semiparametric estimator where no initial input is necessary. The estimation procedure is illustrated via a generalized autoregressive conditional heteroskedasticity (GARCH) model. Asymptotic properties of the estimator are established, and finite sample performance of the estimator is evaluated via simulation. The results suggest that the proposed one-step semiparametric estimator avoids significant drawbacks of its two-step counterparts. (JEL: C02, C22, C51)

When responses are missing at random, we consider semiparametric estimation of inverse density weighted expectations, or equivalently, integrals of conditional expectations. An inverse probability weighted estimator and a full propensity score weighted estimator are proposed and shown to be asymptotically normal. The two estimators are asymptotically equivalent and achieve the semiparametric efficiency bound. The performances of the estimators are investigated and compared with simulation studies and a real data example.

We consider a panel data partially linear single-index models (PDPLSIM) with errors correlated in space and time. A serially correlated error structure is adopted for the correlation in time. We propose using a semiparametric minimum average variance estimation (SMAVE) to obtain estimators for both the parameters and unknown link function. We not only establish an asymptotically normal distribution for the estimators of the parameters in the single index and the linear component of the model, but also obtain an asymptotically normal distribution for the nonparametric local linear estimator of the unknown link function. Then, a fitting of spatial and time-wise correlation structures is investigated. Based on the estimators, we propose a generalized F-type test method to deal with testing problems of index parameters of PDPLSIM with errors correlated in space and time. It is shown that under the null hypothesis, the proposed test statistic follows asymptotically a [Formula: see text]-distribution with the scale constant and degrees of freedom being independent of nuisance parameters or functions. Simulated studies and real data examples have been used to illustrate our proposed methodology.