Search results

The asymptotic bias and variance of a general class of local polynomial estimators of M-regression functions are studied over the whole compact support of the multivariate covariate under a minimal assumption on the support. The support assumption ensures that the vicinity of the boundary of the support will be visited by the multivariate covariate. The results show that like in the univariate case, multivariate local polynomial estimators have good bias and variance properties near the boundary. For the local polynomial regression estimator, we establish its asymptotic normality near the boundary and the usual optimal uniform convergence rate over the whole support. For local polynomial quantile regression, we establish a uniform linearization result which allows us to obtain similar results to the local polynomial regression. We demonstrate both theoretically and numerically that with our uniform results, the common practice of trimming local polynomial regression or quantile estimators to avoid “the boundary effect” is not needed.

Inspired by the basic idea of gradient boosting, this study aims to design a novel multivariate regression ensemble algorithm RegBoost by using multivariate linear regression as a weak predictor.

To achieve nonlinearity after combining all linear regression predictors, the training data is divided into two branches according to the prediction results using the current weak predictor. The linear regression modeling is recursively executed in two branches. In the test phase, test data is distributed to a specific branch to continue with the next weak predictor. The final result is the sum of all weak predictors across the entire path.

Through comparison experiments, it is found that the algorithm RegBoost can achieve similar performance to the gradient boosted decision tree (GBDT). The algorithm is very effective compared to linear regression.

This paper attempts to design a novel regression algorithm RegBoost with reference to GBDT. To the best of the knowledge, for the first time, RegBoost uses linear regression as a weak predictor, and combine with gradient boosting to build an ensemble algorithm.

The purpose of this paper is to provide a new approach for detecting the small sustained process shifts in multistage systems with correlated multiple quality characteristics.

The authors propose a new multivariate linear regression model for a multistage manufacturing system with multivariate quality characteristics in which both the auto-correlated process outputs and the correlations occurring between neighboring stages are considered. Then, the multistage multivariate residual control charts are constructed to monitor the overall process quality of multistage systems with multiple quality characteristics. Moreover, an overall run length concept is adopted to evaluate the performances of the authors’ proposed control charts.

In the numerical example with cascade data, the authors show that the detecting abilities of the proposed multistage residual MEWMA and MCUSUM control charts outperform those of Phase II MEWMA and MCUSUM control charts. It further demonstrates the usefulness of the authors’ proposed control charts in the Phase II monitoring.

The research results of this paper can be applied to any multistage manufacturing or service system with multivariate quality characteristics. This new approach provides quality practitioners a better decision making tool for detecting the small sustained process shifts in multistage systems.

Once the multistage multivariate residual control charts are constructed, one can employ them in monitoring and controlling the process quality of multistage systems with multiple characteristics. This approach can lead to the direction of continuous improvement for any product or service within a company.

The objective of this article is to determine the usage and application of logistic regression analysis in the marketing literature by comparing the market positioning of prominent marketing journals.

In order to identify the logistic regression applications, those journals having “marketing” term in their titles and indexed by the social citation index (SSCI) were included. As a result, the target population consisted of 12 journals fulfilling the criteria set. However, only eight of these that were accessible by the researchers were included in the study.

The classification of marketing articles from the chosen prominent marketing journals were made by journal title, article topic, target population, data collection method, and study location has mapped the position of logistic regression in the marketing literature.

The sample journal coverage was limited with 12 marketing journals indexed in SSCI. In some of the journals utilized, the accessibility was limited by the electronic database year coverage. Due to this limitation, the researchers could not reach the exact number of articles using logistic regression.

The results of this study could highlight what is researched with logistic regression about marketing problems and may shed light on solving different problems on marketing topics for the future.

This paper aims to identify antecedents and consequences of pharmacy loyalty behavior.

A cross-sectional study was conducted. Constructs involved were measured using an online self-administered questionnaire. Data were analyzed using multivariate logistic and linear regression.

In all, 400 usable responses were obtained. General satisfaction (odds ratio [OR] = 1.52; p < 0.01; 95 per cent confidence interval [CI] = 1.12 to 2.06) and trust (OR = 1.81; p < 0.01; 95 per cent CI = 1.32 to 2.50) were found to have statistically significant relationships with loyalty behavior. General satisfaction (regression coefficient = 0.20; p < 0.01; 95 per cent CI = 0.09 to 0.31), explanation component of satisfaction with service quality (regression coefficient = 0.13; p < 0.01; 95 per cent CI = 0.04 to 0.21), consideration and technical competence components of satisfaction with service quality (regression coefficient = 0.18; p = 0.02; 95 per cent CI = 0.03 to 0.33) and trust (regression coefficient = 0.33; p < 0.01; 95 per cent CI = 0.21 to 0.45) were statistically significantly related to positive word-of-mouth promotion. General satisfaction (regression coefficient = −0.29; p < 0.01; 95 per cent CI = −0.3 to −0.18), consideration and technical competence components of satisfaction with service quality (regression coefficient = −0.17; p = 0.02; 95 per cent CI = −0.31 to −0.03) and trust (regression coefficient = −0.21; p < 0.01; 95 per cent CI = −0.33 to −0.10) had statistically significant relationships with negative word-of-mouth promotion.

Pharmacists can utilize these results to develop better marketing strategies. These results can be used by researchers to forward this area of research. This study had some study design limitations that may affect its generalizability.

Effect of satisfaction as a multidimensional construct on pharmacy loyalty behavior and word-of-mouth promotion, identification of drivers of negative word-of-mouth promotion and effect of pharmacy trust on pharmacy loyalty behavior and word-of-mouth promotion are some of the major contributions of this study.

The purpose of this paper is to assess the quality of commercial lubricant oils. A spectroscopic method was used in combination with multivariate regression techniques (ordinary multivariate multiple regression, principal components analysis, partial least squares, and support vector regression (SVR)).

The rationale behind the use of SVR was the fuzzy characteristics of the signal and its inherent ability to find nonlinear, global solutions in highly complex dimensional input spaces. Thus, SVR allows extracting useful information from calibration samples that makes it possible to characterize physical-chemical properties of the lubricant oils.

A dataset of 42 spectra measured from oil standards was studied to assess the concentration of copper into the oils and, thus, evaluate the wearing of the machinery. It was found that the use of SVR was very advantageous to get a regression model.

The use of genetic algorithms coupled to SVR was considered in order to reduce the time needed to find the optimal parameters required to get a suitable prediction model.

The purpose of this paper is to show a comparative study of different direction-of-arrival (DOA) estimation techniques, namely, multiple signal classification (MUSIC) algorithm, delay-and-sum (DAS) beamforming, support vector regression (SVR), multivariate linear regression (MLR) and multivariate curvilinear regression (MCR).

The relative delay between the microphone signals is the key attribute for the implementation of any of these techniques. The machine-learning models SVR, MLR and MCR have been trained using correlation coefficient as the feature set. However, MUSIC uses noise subspace of the covariance-matrix of the signals recorded with the microphone, whereas DAS uses the constructive and destructive interference of the microphone signals.

Variations in root mean square angular error (RMSAE) values are plotted using different DOA estimation techniques at different signal-to-noise-ratio (SNR) values as 10, 14, 18, 22 and 26dB. The RMSAE curve for DAS seems to be smooth as compared to PR1, PR2 and RR but it shows a relatively higher RMSAE at higher SNR. As compared to (DAS, PR1, PR2 and RR), SVR has the lowest RMSAE such that the graph is more suppressed towards the bottom.

DAS has a smooth curve but has higher RMSAE at higher SNR values. All the techniques show a higher RMSAE at the end-fire, i.e. angles near 90°, but comparatively, MUSIC has the lowest RMSAE near the end-fire, supporting the claim that MUSIC outperforms all other algorithms considered.

Value‐at‐Risk (VaR) has become a mainstream risk management technique employed by a large proportion of financial institutions. There exists a substantial amount of research dealing with this task, most commonly referred to as VaR backtesting. A new generation of “self‐learning” VaR models (Conditional Autoregressive Value‐at‐Risk or CAViaR) combine backtesting results with ex ante VaR estimates in an ARIMA framework in order to forecast P/L distributions more accurately. In this commentary, the authors present a systematic overview of several classes of applied statistical techniques that can make VaR backtesting more comprehensive and provide valuable insights into the analytical properties of VaR models in various market environments. In addition, they discuss the challenges associated with extending traditional backtesting approaches for VaR horizons longer than one day and propose solutions to this important problem.

The average U.S. farm size has risen dramatically over the last three decades. Motives for this trend are the subject of a large body of literature. This study incorporates farm size risk and return analysis into this research stream. In this paper, cross-sectional and temporal relations between farm size and returns are examined and characterized.

Relying on farm level panel data from Kansas Farm Management Association (KFMA) for 140 farms from 1996 to 2018, this article examines the relationship between farm size and returns and investigates whether farm size is related to risk. Two measures of farm returns are used: excess return on equity and risk-adjusted return on equity. Value of farm production and total farm acres are used as measures of farm size.

Findings suggest a significant and positive relationship between farm size and excess return on equity as well as farm size and risk-adjusted return on equity. However, this return premium associated with farm size is not associated with additional risk. Stated differently, farm size can be viewed as a farm characteristic that is associated with higher return without additional risk.

These findings provide further support for ongoing farm consolidation.

The results suggest the trend towards consolidation in production agriculture is likely to continue. Larger farms bear less risk.