Principal Component Regression Model Research Articles

Feed shaft thermal expansion error modeling based on principal component regression

In order to further improve the prediction accuracy of the thermal error model of the feed shaft of the gear grinding machine, this paper proposes a feed shaft thermal expansion error modeling method based on principal component regression. The feed shaft positioning error is decoupled by linear fitting to obtain the thermal expansion slope parameter, the position correlation of the feed shaft thermal expansion error is eliminated, and a regression model of the thermal expansion slope parameter and the temperature of all measuring points is established based on the principal component regression algorithm. Different from the traditional method, the principal component regression model does not require additional screening of temperature sensitive points, and the root mean square error mean and standard deviation of the group experimental prediction results can reach 2.0 μm/m、0.9μm/m, which has higher accuracy and stability than conventional methods.

Material-Sparing Approach to Predict Tablet Capping Under Processing Compression Conditions Based on Mechanical and Molecular Properties Derived from Compaction Simulation and Crystal Structural Analysis.

Present study evaluates the usability of compaction simulation-based mechanical models as a material-sparing approach to predict tablet capping under processing compression conditions using Acetaminophen (APAP) and Ibuprofen (IBU). Measured mechanical properties were evaluated using principal component analysis (PCA) and principal component regression (PCR) models. PCR models were then utilized to predict the capping score (CS) from compression pressure (CP). APAP formulations displayed a quadratic correlation between CS and CP, with CS rank order following CP of 200MPa < 300MPa < 100MPa, indicating threshold compression pressure (TCP) limit between 200 and 300 MPa, resulting in higher CS at 300 than 200 MPa regardless of increased CP. IBU formulations displayed a linear correlation between CS and CP, with CS rank order following CP of 100MPa < 200MPa < 300MPa, indicating TCP limit between 100 and 200 MPa, resulting in higher CS at 200 and 300 than 100 MPa regardless of increased CP. Molecular models were developed as validation methods to predict capping from CP. Measured XRPD patterns of compressed tablets were linked with calculated Eatt and d-spacing of slip planes and analyzed using variable component least square methods to predict TCP triggering cleavage in slip planes and leading to capping. In APAP and IBU, TCP values were predicted at 245 and 175 MPa, meaning capped tablets above these TCP limits regardless of increased CP. A similar trend was observed in CS predictions from mechanical assessment, confirming that compaction simulation-based mechanical models can predict capping risk under desired compression conditions rapidly and accurately.

Predicting water content in engine oil using SPA-ISSA-BP methods &amp;nbsp;

Contamination of engine oil with water can accelerate oil oxidation and decomposition, resulting in oil degradation. Therefore, the detection and prediction of water content in engine oil is crucial to ensure the long-term safe operation of engines. In this study, near-infrared spectroscopy was used to analyse engine oils with varying water contents. Spectral data were obtained, and various methods, including Savitzky-Golay (SG) smoothing, orthogonal signal correction (OSC), and successive projection algorithm (SPA) were employed to compare the performance differences of principal component regression models. An Improved Sparrow Searches Algorithm (ISSA) was then applied to optimise the Back-Propagation Neural Network (BPNN) for predicting water content in SAE 10W-30 engine oil. The results showed that the performance of the principal component regression model constructed from spectral data after SG smoothing, OSC, and the SPA feature wavelength selection had improved. The BPNN optimised by the improved Sparrow Search Algorithm accelerated the convergence speed of the model and effectively improved the prediction accuracy of the BPNN. The obtained coefficient of determination (R2) was 0.99103, the root mean square error (RMSE) was 2. 2136 10-4, and the mean absolute error (MAE) was 1. 8889×10-4. These results provide an effective method for detecting and predicting the water content in engine oil.

Effect of spatial differentiation of Pinus massoniana canopy branches and leaves on rainfall interception – an example from Changting County

ABSTRACT The spatial differentiation of forest canopy branches and leaves affects rainfall interception. However, the scientific expression of the spatial differentiation of forest canopy branches and leaves has not been clarified, and revealing the effect of spatial differentiation of branches and leaves of the vertical structure of the forest canopy on rainfall interception is difficult. Using the Pinus massoniana planted in Changting County, China, as the research object, this study quantified the spatial differentiation of canopy branches and leaves in a hierarchical manner by combining the unmanned aerial vehicle (UAV) three-dimensional point cloud model with the landscape metrics and further simulated the interception mechanism of Pinus massoniana forest canopy on rainfall of different intensities by constructing the principal component regression model. The results showed that the spatial differentiation of canopy branches and leaves in different strata responds differently to different rainfall intensities.

Prediction of freezing point and moisture distribution of beef with dual freeze-thaw cycles using hyperspectral imaging

The freezing point (FP) is an important quality indicator of the superchilled meat. Currently, the potential of hyperspectral imaging (HSI) for predicting beef FP as affected by multiple freeze-thaw (F-T) cycles was explored. Correlation analysis revealed that the FP had a negative correlation with the proportion of bound water (P21) and a positive correlation with the proportion of immobilized water (P22). Moreover, the optimal wavelengths were selected by principal component analysis (PCA). Principal component regression (PCR) and partial least squares regression (PLSR) models were successfully developed based on the optimal wavelengths for predicting FP with determination coefficient in prediction (RP2) of 0.76, 0.76 and root mean square errors in prediction (RMSEP) of 0.12, 0.12, respectively. Additionally, PLSR based on full wavelengths was established for predicting P21 with RP2 of 0.80 and RMSEP of 0.67, and PLSR based on the optimal wavelengths was established for predicting P22 with RP2 of 0.87 and RMSEP of 0.66. The results show the potential of hyperspectral technology to predict the FP and moisture distribution of meat as a nondestructive method.

Functional Principal Component Regression Model Based on FIrst-order Boosting and Model Averaging

Functional regression model has always been an important research direction in the field of functional data analysis, in which functional principal component analysis as a common method of functional regression modeling, the determination of the number of truncations of the eigenfunctions has always been a problem to be solved. Based on this, this paper proposes a functional principal component regression model based on the idea of model averaging, which uses Bayesian model averaging instead of model selection, determines the number of truncations adaptively and balances the prediction error with the training error. In addition, the original function curves do not effectively capture the information of their functional characteristics. In this paper, the first-order derivative function is used as the supplementary information of the original function curve to jointly characterize the predictor variables. In the efficiency test, this paper averages the Bayesian model as the benchmark model and conducts real data analysis on several benchmark datasets. The results show that the proposed method has higher predictive accuracy and robustness than commonly used models.

Forecasting wine phenolic composition from infrared spectra of grapes extracts and monitoring of fermentations with optimised time-specific prediction models

Incorporating monitoring and forecasting technologies has the potential to enhance the different aspects of winemaking. By monitoring certain chemical parameters, it is possible to evaluate the progress of the fermentation and phenolic extraction, while also detecting deviations from expected behaviour. This study aims to investigate the development of a rapid forecasting method and different strategies to improve the accuracy of process monitoring. The first part involves the introduction of partial least squares (PLS) calibration models able to forecast the trajectories of phenolic extraction during fermentation using grape extract spectra. Additionally, novel methods for reoptimizing calibration models are proposed, with the intention of increasing the prediction accuracy. Whilst this method is transferable to PLS models, principal component regression (PCR) models were used instead as a more simplistic way to test the new methods. First, dynamic PCR linear regression coefficients were adapted over time to capture the compound evolution in the wine matrix. Secondly, PCR calibration models were reoptimized using interpolated spectral and reference data. The forecasting PLS calibration models have shown promising results for total anthocyanin content (mg/L), colour density, polymeric pigments (mg/L), MCP tannins (mg/L), and total phenolic index with overall R2 values on the validation set above 0.6, and low RSMEV values. Using interpolated data to re-optimise models was found to be more effective than global models or those using the calculated regression coefficients methods for predicting the concentration of a particular compound at a certain time point. While R2 values are suboptimal for all considered PCR models, in 72% of the cases analysed, the interpolation method for re-optimisation showed RMSEV values between 6% and 83% lower than those reported for a more traditional PCR model and other re-optimisation strategies. Interpolated models have shown consistency in predicting values, along with good performance metrics, indicating that they could be applied to samples for which calibration models do not exist.

China’s inflation forecasting in a data-rich environment: based on machine learning algorithms

ABSTRACT Inflation forecasting stands as a central concern in macroeconomics. This paper focuses on predicting China’s inflation within a data-rich environment. Specifically, we compile a large panel of China’s monthly macroeconomic and financial variables, employing various machine learning models on this predictor panel to forecast China’s inflation, encompassing both CPI and PPI, across various forecasting horizons extending up to 24 months. Our findings indicate the following: First, the machine learning models, when coupled with a large dataset of macroeconomic and financial predictors, demonstrate a superior ability to forecast China’s inflation compared to benchmark time series models and principal component regression models. These advantages become even more notable at medium-to-long horizons and during periods of high inflation volatility, as well as CPI/PPI divergence periods. Second, our experiments reveal that penalized linear regression models, such as ridge and elastic net, consistently outperform the benchmarks and nonlinear machine learning methods in most cases. Lastly, variables related to price, stock, and money & credit are identified as the most crucial factors for forecasting inflation in China.

Risk governance and regulatory adjustments in the public commercial banks of OECD

PurposeThis study aims to explore the relationship between risk governance characteristics (chief risk officer [CRO], chief financial officer [CFO] and senior directors [SENIOR]) and regulatory adjustments (RAs) in Organization for Economic Cooperation and Development public commercial banks.Design/methodology/approachUsing principal component analysis (PCA) and regression models, the research analyzes a representative data set of these banks.FindingsA significant negative correlation between risk governance characteristics and RAs is found. Sensitivity analysis on the regulatory Tier 1 capital ratio and the total capital ratio indicates mixed outcomes, suggesting a complex relationship that warrants further exploration.Research limitations/implicationsThe study’s limited sample size calls for further research to confirm findings and explore risk governance’s impact on banks’ capital structures.Practical implicationsEnhanced risk governance could reduce RAs, influencing banking policy.Social implicationsThe study advocates for improved banking regulatory practices, potentially increasing sector stability and public trust.Originality/valueThis study contributes to understanding risk governance’s role in regulatory compliance, offering insights for policymaking in banking.

The Positive Association between Grit and Mental Toughness, Enhanced by a Minimum of 75 Minutes of Moderate-to-Vigorous Physical Activity, among US Students

Drawing from the 2015 Gucciardi et al.’s mental toughness (MT) framework, this study examines the association between grit and MT in US college students, while considering the moderating role of at least 75 min of moderate-to-vigorous physical activity (MVPA) based on recommendations from the American College of Sports Medicine. We administered the Grit-S Scale and the Mental Toughness Index in two samples of a total of 340 US undergraduate student-athletes and graduate students. The Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies was employed to ensure internal validity, while statistical procedures including principal component analysis and regression models were utilized to analyze the collected data. A weighted component combining grit and the interaction between MVPA and grit significantly predicted MT, explaining 23% of its variability. Drawing from a specific conceptual framework, this study provides novel insights into the relationship between grit, engagement in at least 75 min of MVPA per week, and MT among US collegiate students. The findings support a positive association between grit, MVPA, and both MT and a specific component of MT, highlighting the significance of these factors in enhancing performance and suggesting potential implications for future research and practical applications in the field.

Transformations for non-destructive evaluation of brix in mango by reflectance spectroscopy and machine learning

Mango is a very popular climacteric fruit in America and Europe. Among the internal properties of the mango, total soluble solids (TSS) are an adequate indicator to estimate the quality of mango, however, the measurement of this indicator requires destructive tests. Several research have addressed similar issues; they have made use of pre-processing transformations without making it clear which of them is statistically better. Here, we created a new spectral database to build machine learning (ML) models. We analyzed a total of 18 principal component regression (PCR) models and 18 partial least squared regression (PLSR) models, where 4 types of transformations, 3 different feature extractors, and 3 different pre-processing techniques are combined. The research proposes a double cross validation (CV) both to determine the optimal number of components and to obtain the final metrics. The best model had a root mean square error (RMSE) of 1.1382 °Brix and a RMSE on the transformed scale of 0.5140. The best model used 4 components, used y&lt;sup&gt;2&lt;/sup&gt; transformation, reflectance R as the independent variable and MSC as a pre-processing technique.

Bayesian LASSO for population stratification correction in rare haplotype association studies.

Population stratification (PS) is one major source of confounding in both single nucleotide polymorphism (SNP) and haplotype association studies. To address PS, principal component regression (PCR) and linear mixed model (LMM) are the current standards for SNP associations, which are also commonly borrowed for haplotype studies. However, the underfitting and overfitting problems introduced by PCR and LMM, respectively, have yet to be addressed. Furthermore, there have been only a few theoretical approaches proposed to address PS specifically for haplotypes. In this paper, we propose a new method under the Bayesian LASSO framework, QBLstrat, to account for PS in identifying rare and common haplotypes associated with a continuous trait of interest. QBLstrat utilizes a large number of principal components (PCs) with appropriate priors to sufficiently correct for PS, while shrinking the estimates of unassociated haplotypes and PCs. We compare the performance of QBLstrat with the Bayesian counterparts of PCR and LMM and a current method, haplo.stats. Extensive simulation studies and real data analyses show that QBLstrat is superior in controlling false positives while maintaining competitive power for identifying true positives under PS.

Rapid monitoring of heavy metal-ammonia complexes in solutions by UV–vis/ATR-FTIR spectroscopy and chemometric models

Ammonia can form stable chemical complexes with heavy metal ions, which significantly affect the treatment efficiency of these wastewaters. Thus, accurately determining the concentration of heavy metal ammonia complex ions is crucial to understanding the complex decomposition mechanism. The current study predicted the composition of copper ammonia solution through thermodynamic calculations. The mid-infrared first derivative spectra obtained from density functional theory (DFT) calculation and measurements of the simulated solutions were compared, which indicated that [Cu(NH3)4]2+ was the primary form of complexes in solution. The acquired FTIR spectra and UV–vis spectra demonstrated a strong correlation with the [Cu(NH3)4]2+ concentrations; whereas a principal component regression (PCR) model was implemented for concentration computation. The UV–vis-PCR model was found to be a fast and accurate method for quantification of copper-ammonia complex concentration, and its detection limit and quantification limit could reach 5 mg·L−1 and 10 mg·L−1, respectively, in the 5–200 mg·L−1 range. This study established qualitative and quantitative methods for analyzing copper ammonia complex in water, which provided the feasibility for rapid real-time monitoring of copper ammonia complex.

Differentiation through E-nose and GC-FID data modeling of rosé sparkling wines elaborated via traditional and Charmat methods.

The growing demand for rosé sparkling wine has led to an increase in its production. Traditional or Charmat wine-making influence the aromatic profiles in wine. An analysis such as gas chromatography makes an accurate assessment of wines based on volatile detection but is resource intensive. On the other hand, the electronic nose (E-nose) has emerged as a versatile tool, offering rapid, cost-effective discrimination of wines, and contributing insights into quality and production processes because of its aptitude to perform a global aromatic pattern evaluation. In the present study, rosé sparkling wines were produced using both methods and major volatile compounds and polyols were measured. Wines were tested by E-nose and predictive modelling was performed to distinguish them. Volatile profiles showed differences between Charmat and traditional methods, especially at 5 months of aging. A partial least square discriminant analysis (PLS-DA) was carried out on E-nose detections, obtaining a model that describes 94% of the variability, separating samples in different clusters and correctly identifying different classes. The differences derived from PLS-DA clustering agree with the results obtained by gas-chromatography. Moreover, a principal components regression model was built to verify the ability of the E-nose to non-destructively predict the amount of different volatiles analyzed. Production methods of Rosé sparkling wine affect the final wine aroma profiles as a result of the differences in terms of volatiles. The PLS-DA of the data obtained with E-nose reveals that distinguishing between Charmat and traditional methods is possible. Moreover, predictive models using gas chromatography-flame ionization detection analysis and E-nose highlight the possibility of fast and efficient prediction of volatiles from the E-nose. © 2023 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Comparison of principal component regression (PCR) and partial least square regression (PLSR) modeling methods for quantifying polyethylene (PE) in recycled polypropylene (rPP) with near-infrared spectrometry (NIR)

Recycled polypropylene (rPP) often contains a small amount of polyethylene (PE). Since polypropylene (PP) and PE are incompatible, the presence of PE compromises the performance of rPP materials and needs to be closely monitored. In our previous work, Raman and near-infrared (NIR) spectrometries were evaluated to monitor PE content in rPP with partial least square regression (PLSR) modeling. The NIR spectrometry exhibited a wider application range, but the accuracy of the prediction models might be further improved. In the current work, a different modeling method, principal component regression (PCR) was employed to analyze PE content in rPP with NIR spectrometry. Spectrum pretreatment methods, including multivariate scatter correction (MSC), standard normal variate transformation (SNV), smoothing, and first derivative, were investigated to improve the NIR spectrum quality. Forward and backward interval methods were used to optimize spectral range selection. The outcomes were compared with our previous PLSR modeling results. The highest accuracy in independent validation was achieved by a PCR model with an R 2 of 0.9991 and a root-mean-square error of prediction (RMSEP) of 0.1596 PE%. On the other hand, a PLSR model achieved the lowest RMSEP of 0.9712 PE% for a non-colored post-consumer rPP sample. The PCR models might be sensitive to interference and more suitable for post-industrial materials, which have a simpler chemical composition. The PLSR models might have better stability and be more suitable for complicated post-consumer samples. Both the PCR and PLSR models were successfully applied to a gray commercial rPP sample.

Analysis of the geotechnical and mineralogical characteristics of the Settat-Khouribga shale clay for potential civil engineering applications

This study provides a comprehensive analysis of clay shale from the Settat-Khouribga region, assessing its suitability for civil engineering applications through various tests. Grain size distribution, hardness (MDE, LA), and plasticity (ES 0/5, IP, VB) tests, along with consistency index evaluations following the "CSTCN" catalog, were conducted. X-ray diffraction (XRD) and mineralogical analyses identified key components such as quartz, calcite, feldspar, and iron oxide, with quartz being predominant. The results demonstrated that the shale clay is well-suited for construction purposes, exhibiting favorable grain size distribution and high resistance to wear and abrasion, which are crucial for material durability. Its low plasticity further indicates good mechanical stability in construction projects. SEM analysis revealed a microstructure dominated by quartz grains, montmorillonite, and illite lamellae, and the presence of micro fissures and pores suggests both strengths and potential durability concerns. Advanced statistical analyses, including Principal Component Analysis (PCA) and regression models, were employed to assess geotechnical properties. The study confirms the shale's compliance with construction standards, highlighting its potential for sustainable building practices. These findings contribute significantly to the field of geotechnical engineering, emphasizing the material's suitability for construction and its broader societal implications.

Application of GIS and spatiotemporal analyses in viral infection modelling using multiple datasets – A case study on the SARS-CoV-2 epidemic

Introduction/objectiveViral and infectious diseases such as COVID-19 continue to pose a significant public health threat. In order to create an early warning system for new pandemics or emerging versions of the virus, it is imperative to study its epidemiology. In this study, we created a geospatial model to predict the weekly contagion and lethality rates of COVID-19 in Ireland. MethodsMore than forty parameters including atmospheric pollutants, metrological variables, sociodemographic factors, and lockdown phases were introduced as input variables to the model. The significant parameters in predicting the number of new cases and the death toll were identified. QGIS software was employed to process input data, and a principal component regression (PCR) model was developed using the statistical add-on XLSTAT. Results and conclusionsThe developed models were able to predict more than half of the variations in contagion and lethality rates. This indicates that the proposed model can serve to help prediction systems for the identification of future high-risk conditions. Nevertheless, there are additional parameters that could be included in future models, such as the number of deaths in care homes, the percentage of contagion and mortality among health workers, and the degree of compliance with social distancing.

Association of Geographic Distance and Hospital Characteristics With Use of Interhospital Transfer by Air: A Multicenter Retrospective Study

ObjectiveInterhospital transfer by air (IHTA) represents the majority of helicopter air ambulance transports in the United States, but the evaluation of what factors are associated with utilization has been limited. We aimed to assess the association of geographic distance and hospital characteristics (including patient volume) with the use of IHTA. MethodsThis was a multicenter, retrospective study of helicopter flight request data from 2018 provided by a convenience sample of 4 critical care transport medicine programs in 3 US census regions. Nonfederal referring hospitals located in the home state of the associated critical care transport medicine program and within 100 miles of the primary receiving facility in the region were included if complete data were available. We fit a Poisson principal component regression model incorporating geographic distance, the number of emergency department visits, the number of hospital discharges, case mix index, the number of intensive care unit beds, and the number of general beds and tested the association of the variables with helicopter emergency medical services utilization. ResultsA total of 106 referring hospitals were analyzed, 21 of which were hospitals identified as having a consistent request pattern. Using the hospitals with a consistent referral pattern, geographic distance had a significant positive association with flight request volume. Other variables, including emergency department visit volume, were not associated. Overall, the included variables offered poor explanatory power for the observed variation between referring facilities in the use of IHTA (r2 = 0.09). Predicted flights based on the principal component regression model for all referring hospitals suggested the majority of referring hospitals used multiple flight programs. ConclusionGeographic distance is associated with the use of IHTA. Unexpectedly, most basic hospital characteristics are not associated with the use of IHTA, and the degree of variation between referring facilities that is explained by patient volume is limited. The evaluation of nonhospital factors, such as the density and availability of critical care or advanced life support ground emergency medical services resources, is needed.

Mineral oil detection in ghee using attenuated total reflectance-fourier transform infrared spectroscopy (ATR-FTIR) in conjunction with chemometrics

Ghee (anhydrous milk fat), being a costly product, is susceptible to adulteration by unscrupulous personnel in the market. The presence of mineral oil in ghee was examined using Attenuated Total Reflectance-Fourier Transform Infrared spectroscopy (ATR-FTIR) spectroscopy in conjunction with chemometrics. Spectral data in the wavenumber range 4000–500 cm−1 were collected for pure ghee, mineral oil, and pure ghee samples spiked with mineral oil at different concentrations (2 %, 4 %, 6 %, 10 %, and 15 %). Principal Component Analysis (PCA) revealed distinct clustering patterns in specific wavenumber regions (1350–950 and 1800–1600 cm−1) at 5% level of significance, effectively distinguishing pure ghee from adulterated samples. Soft Independent Modeling of Class Analogy (SIMCA) showed classification efficiency of 100 % for the samples of pure ghee and mineral oil. Developed Partial Least Square (PLS) and Principal Component Regression (PCR) models showed high R2 values (above 0.99) indicating strong predictive accuracy for mineral oil adulteration levels in ghee. Thus, using ATR-FTIR and chemometrics, detection of as low as 2 % of mineral oil in ghee is possible.

Depressive symptoms associated with brain morphometry in Mild Cognitive Impairment stage due to Alzheimer’s and Parkinson’s disease

Objectives: Although there are studies linking depressive symptoms with psychiatric diseases, and then, with brain morphometry, scarce is the literature examining the association between depressive symptoms and brain morphometry in the prodromal stage of neurodegenerative disorders. In the current analysis, we used a fair sample of Mild Cognitive Impairment (MCI) stage due to Alzheimer’s disease (AD) and Parkinson’s disease (PD) patients, to examine the association between different depressive symptoms and a large variety of brain morphometric factors.&#x0D; Methods/ Design: Our sample consisted of 352 patients. The Geriatric Depression Scale (GDS) was used for the association with brain measurements. Principal Components Analysis (PCA) and regression models were used for the analyses. Age, sex, years of education, Addenbrooke’s cognitive examination revised (ACE-R), and estimated total intracranial volume were used as covariates.&#x0D; Results: Factor3 was significantly and positively correlated with GDS total, Dysthymia, Anxiety and Hopelessness. r_factor1, r_factor2 and r_factor5 were significantly and positively correlated with Anxiety, while r_factor5 was also correlated with GDS total. Concerning volumes, v_factor3 was found to be correlated with GDS total, Dysthymia, Anxiety and Hopelessness.&#x0D; Conclusions: The association between depressive symptoms and upcoming neurodegenerative disorders is reflected in human brain structural alterations, suggesting specific modifications in network-like patterns and neuronal connectivity. Characterizing differences in the morphometry of brain regions among MCI patients with depressive symptoms may aid the early diagnosis and treatment of these symptoms.