Prediction Deviation Research Articles

Prediction of Moisture Content in Gambir Leaf Herbal Tea Using NIR Spectroscopy

Abstract Gambir leaf herbal tea is an innovative product made from Gambir (Uncaria gambir Roxb.), which can potentially be an export product. Considering its hygroscopic nature, moisture content is an imperative indicator for controlling the stability of Gambir leaf herbal tea quality. This research intends to investigate near-infrared spectroscopy (NIR) to estimate the moisture content of herbal tea made from Gambir leaves. We placed powdered Gambir leaf herbal tea samples in a petri dish, using 20-21 g per sample. The number of samples used was 60. We obtained the reflectance of herbal tea from Gambir leaves using NIR spectroscopy at 1000 to 2500 nm. The oven method determined the reference moisture content. A calibration model was developed using the partial least squares (PLS) regression method. Furthermore, the model acquires standard error in the calibration (SEC) and prediction (SEP) are 0.36 and 0.43, respectively. The model had a determination coefficient of prediction set (R2p) of 0.95 and Residual Predictive Deviations (RPD) of 4.38. This study proves that NIR spectroscopy is reliable in presupposing the moisture content of herbal tea made from Gambir leaves.

Numerical study of the effects of surface nitrogen-containing on soot formation process in ammonia/ethylene counterflow diffusion flames

In this paper, a modified soot kinetic model was developed to explore the potential inhibitory effects of nitrogen-containing functional groups on the soot surface growth process. In the modified soot kinetic model, the formation process of cyano functional groups on the soot surface was simulated through the reaction of hydrogen cyanide (HCN) and the surface open sites, and the surface cyano functional groups were assumed to no longer participate in the subsequent hydrogen-abstraction-acetylene-addition (HACA) reaction. The prediction performance of soot volume fraction (SVF) and average particle size (D63) in ammonia doping ethylene counterflow diffusion flames was compared between the traditional soot kinetic model and the modified kinetic soot model. The results show that both models reflected the suppression effects of ammonia doping on soot formation observed in the experiments. However, compared to the traditional soot kinetic model, the modified soot kinetic model exhibited better performances in predicting SVF and D63 with prediction deviations reduced by 52.7 % and 2.3 nm, respectively. The simulation results showed the formation of surface cyano functional groups led to a nitrogen content of up to 0.84 % in the soot, which is consistent with previous experimental measurements. In-depth analysis shows that the formation of cyano functional groups on the soot surface decreased the proportion of hydrogen sites and further diminished the number of open sites, resulting in reductions in the rates of HACA surface growth and the soot mass growth by up to 22 % and 29 %, respectively. It is proposed that the nitrogen-containing functional groups on the surface of soot in ammonia-doped hydrocarbon flames inhibit the formation of soot, providing an alternative pathway to the well-known gas-phase suppression mechanism.

Decoding the synergy: unveiling gradient boosting regression model for multivariate quantitation of pioglitazone, alogliptin and glimepiride in pure and tablet dosage forms

This study represents a comparison among the performances of four multivariate procedures: partial least square (PLS) and artificial neural networks (ANN) in addition to support vector regression (SVR) and extreme gradient boosting (XG Boost) algorithm for the determination of the anti-diabetic mixture of pioglitazone (PIO), alogliptin (ALG) and glimepiride (GLM) in pharmaceutical formulations with aid of UV spectrometry. Key wavelengths were selected using knowledge-based variable selection and various preprocessing methods (e.g., mean centering, orthogonal scatter correction, and principal component analysis) to minimize noise and improve model precision. XG Boost effectively enhanced computing speed and accuracy by focusing on specific spectral features rather than the entire spectrum, demonstrating its advantages in resolving complex, overlapping spectral data. The independent test results of different models demonstrated that XG Boost outperformed other methods. XG Boost achieved the lowest root mean squared error of prediction (RMSEP) and standard deviation (SD) values across all compounds, indicating minimal prediction error and variability. For PIO, XG Boost recorded an RMSEP of 0.100 and SD of 0.369, significantly better than PLS and ANN. For ALG, XG Boost showed near-perfect performance with an RMSEP of 0.001 and SD of 0.005, outperforming SVR and PLS, which had higher error rates. In the case of GLM, XG Boost also excelled with an RMSEP of 0.001 and SD of 0.018, demonstrating superior precision compared to the much higher errors seen in PLS and ANN. These results highlight XG Boost’s exceptional ability to handle complex, overlapping spectral data, making it the most reliable and accurate model in this study.

Quantitative detection of winter wheat heading phenology based hyperspectral vegetation indexes

The variability of heading rate has considered crucial for quantifying winter wheat heading phenological stage. However, little information is available so far to make a systematic analysis on the quantitative relationship of heading rate to hyperspectral vegetation indexes. Therefore, the establishment of a vegetation index that is sensitive to the changes of heading rate is important for monitoring heading phenology through hyperspectral remote-sensing data. In this study, field experiment was conducted under different cultivation conditions (varied variety, sowing date) at three wheat eco-sites in Shaanxi Province, China. The field canopy reflectance spectrum was synchronously determined with the heading rate within 0.5 m2 area at the heading stage of winter wheat. Newly developed and published spectral indexes were calculated to estimate winter wheat heading rate. Results showed that the newly developed three-band index (R 1227 -R 1236 )/(R 688 -R 1236 ) has demonstrated potential for monitoring the heading rate of winter wheat. This index was derived from a combination of reflectance values at three specific wavelengths: 688 nm, 1227 nm, and 1236 nm, generated the coefficients of determination, root mean square error, and residual prediction deviation values between the measured and predicted values of 0.71, 21.13%, and 1.68, respectively, which is the best indicator for monitoring winter wheat heading rate. This study suggests that winter wheat heading phenology could be accurately estimated based on hyperspectral vegetation indexes. This work has great potential to provide valuable support for assessing winter wheat heading status and proving precision management strategy.

Enhanced dynamic modeling of regenerative CO2 transcritical power cycles: Comparative analysis of Pham-corrected and conventional turbine models

Turbine modeling plays a critical role in assessing system power output and performance in CO₂ transcritical power cycle (CTPC) systems, particularly under dynamic operating conditions. Conventional turbine models often fail to provide accurate predictions when inlet parameters deviate from the design point, leading to discrepancies in system behavior. This paper introduces a novel methodology by integrating a Pham-based corrected turbine performance map into a regenerative CTPC dynamic system. This approach enhances turbine modeling accuracy compared to traditional models, such as the uncorrected turbine performance map, nozzle model, and empirical model. The proposed Pham-corrected model method improves system response stability, particularly in terms of inlet pressure and CO₂ mass flow rate under heat source temperature reductions, offering more reliable performance predictions. In contrast, the empirical model shows substantial deviations in net power output (32 kW or 4.0 %) and thermal efficiency (1.36 %), leading to longer system stabilization times. The Pham-corrected and uncorrected turbine performance map models exhibit similar responses under cooling water mass flow rate reductions but diverge in turbine inlet pressure trends compared to the nozzle and empirical models. Additionally, the uncorrected turbine performance map model shows significant deviations in net power (25.6 kW or 3.17 %) and thermal efficiency (0.5 %) under a 10 % reduction in pump speed. This study highlights the importance of correcting turbine inlet parameters and addressing prediction deviations in conventional models. The Pham-corrected turbine model improves reliability, performance prediction, and design optimization, especially under fluctuating heat sources. With its potential to enhance system stability and efficiency, this methodology offers significant prospects for future CTPC applications, including waste heat recovery and renewable energy integration.

Feature Variable Selection for Near‐Infrared Spectroscopy Based on Simulated Annealing Bee Colony Algorithm

ABSTRACTVariable selection is an effective method to enhance the modeling performance of near‐infrared spectroscopy. Given the promising application prospects of intelligent optimization algorithms in spectral feature variable selection, this article combines the artificial bee colony algorithm with the simulated annealing algorithm to construct a simulated annealing bee colony algorithm (SABC). To explore the feasibility of SABC for spectral variable selection, SABC was applied to construct a partial least squares spectral quantitative detection model for corn stover cellulose and soil organic matter contents. The modeling performance was compared with that of the full spectrum, genetic algorithm, simulated annealing algorithm, and artificial bee colony algorithm; it was found that the model regression precision established by SABC was the best. For the cellulose and organic matter content detection models, the coefficients of determination of the validation set were 0.9433 and 0.9853, with the relative root mean squared error of 1.7901% and 0.8011%, and the residual prediction deviation of 4.1741 and 8.3931, respectively, which could meet the corresponding actual detection needs. SABC adopted the strategy of multiple runs to select the repeated wavelength variables, effectively reduced variable dimensions and model complexity, improved the prediction performance of the regression model, and provided a new approach for building a high‐performance near‐infrared spectroscopy (NIRS) quantitative calibration model.

Identification and crude protein prediction of porcini mushrooms via deep learning-assisted FTIR fingerprinting

Wild edible mushrooms are natural, green, and sustainable foods, and the key to their post-harvest safety and quality control lies in species identification, geographic traceability, and quality assessment. This study analyses the validity of partial least squares discriminant analysis (PLS-DA) model with residual convolutional neural network (ResNet) model based on Fourier transform infrared (FTIR) spectroscopy and two-dimensional correlation spectroscopy (2DCOS) for identifying species and geographic origin of porcini mushrooms. Exploring the feasibility of the partial least squares regression (PLSR) model and long short-term memory network (LSTM) model for predicting the crude protein content of porcini mushrooms. The results show that the ResNet model outperforms PLS-DA with 1.00 and 0.98 prediction results and 1.00 and 0.92 prediction accuracy for new samples, respectively. The LSTM model was more effective than the PLSR model in estimating crude protein content. Among them, LSTM model with first order derivative-multiple scattering correction-Savitzky-Golay (1D-MSC-SG) preprocessing combination was optimal with a prediction set RPD = 2.7, Sig. = 0.418. Besides, the sequential projection algorithm (SPA) feature extraction improves the prediction ability of the LSTM model. Its residual prediction deviation (RPD) was 2.9, Sig. = 0.627. The study provides a new reference for quality evaluation of edible mushrooms and other food products in the market.

Near Infrared Reflectance Spectroscopy-Driven Chemometric Modeling for Predicting Key Quality Traits in Lablab Bean (Lablab purpureus L.) Germplasm

Lablab bean (Lablab purpureus L.) is a multipurpose crop, commonly used for food, feed, and fodder, and its potential as a plant-based meat alternative. Its nutritional diversity, including high protein, starch, and phenolic content, makes it a suitable candidate for nutritional profiling, which is essential for developing nutritionally enhanced varieties. Traditional methods for analyzing its nutritional parameters are labor-intensive, time-consuming, and expensive. This study employs Near-Infrared Reflectance Spectroscopy (NIRS) as a rapid, non-destructive alternative to evaluate 112 Lablab bean genotypes. We developed prediction models for starch, amylose, protein, fat, and phenols using a Modified Partial Least Squares (MPLS) approach, with spectral pre-processing using Standard Normal Variate (SNV) to remove scatter effects and Detrending (DT) to reduce baseline shifts and noise. The models were optimized for derivatives, gap selection, and smoothing, and evaluated using independent test data and key performance metrics including coefficient of determination (R²), bias, and Residual Prediction Deviation (RPD). The best-performing models were: starch (R² = 0.959, RPD = 4.57), amylose (R² = 0.737, RPD = 1.76), protein (R² = 0.911, RPD = 3.09), fat (R² = 0.894, RPD = 2.92), and phenols (R² = 0.816, RPD = 2.36). Statistical tests, including paired t-tests, correlation, and reliability analysis, confirmed the robustness of these models. This study presents a first report offering rapid, multi-trait assessment method for evaluating Lablab bean germplasm, demonstrating high predictive accuracy for pre-breeding practices. It has broad applications in developing nutritionally enhanced varieties, supporting plant-based protein alternatives, and optimizing food production processes to meet the growing demand for healthier, sustainable foods.

Combination of near infrared spectroscopy with characteristic interval selection for rapid detection of rice protein content

Protein level significantly influences the nutritional quality of rice. For this reason, this study introduced a method to rapidly measure the rice protein content through a combination of near infrared spectroscopy (NIRS) with characteristic spectral interval (CSI) selection. Using the interval partial least squares (iPLS) concept as a basis, this study integrated genetic simulated annealing algorithm (GSA) with partial least squares (PLS) and support vector machine (SVM) to develop two CSI selection algorithms, namely GSA-iPLS and GSA-iSVM, respectively. The CSI selected by the above algorithms were compared with synergy iPLS and backward iPLS, and quantitative calibration models were established for PLS and SVM, respectively. The study revealed that the PLS calibration model for rice protein content, developed using CSI selected by GSA-iPLS, exhibited the highest regression accuracy. The optimal model achieved determination coefficients of 0.945 and 0.964, relative root mean square errors of 2.598 % and 2.796 %, and residual predictive deviations of 4.265 and 5.023 for the validation and the external test sets, respectively, which met practical detection requirements. The results indicate that the combination NIRS with GSA CSI intelligent search is a reliable approach for the rapid and accurate detection of rice protein content.

Attenuated Total Reflectance Crystal of Silicon for Rapid Nitrate Sensing Combining Mid-Infrared Spectroscopy.

Rapid detection of NO3 --N is critical to address the challenges of food security, environmental degradation, and climate change. Conventional methods for sensing NO3 --N in water demand pretreatments and chemical reagents, which are time- and cost-consuming. Consequently, Fourier transform infrared attenuated total reflectance (FTIR-ATR) spectroscopy has been well applied for the determination of NO3 --N. However, the conventional ATR crystals, i.e., zinc selenide (ZnSe) and diamond, showed a weakness in duration or cost since the ZnSe material was relatively soft and diamond was relatively expensive. In this study, comparing with ZnSe-ATR and diamond-ATR, a silicon-based ATR (Si-ATR) accessory was developed and used to explore the applicability and stability for sensing NO3 --N combining mathematic algorithms. It was found that partial least-squares regression (PLSR) showed a good performance comparing with the algorithms of principal component analysis (PCA) and linear regression (LR), and it was recommended for quantifying NO3 --N. For ZnSe-ATR, the residual prediction deviation (RPD) was more than 1.80, the determination coefficient (R 2) was more than 0.7725, and the root-mean-square error (RMSE) was less than 2.73 mg L-1. For diamond-ATR, the RPD was more than 1.76, the R 2 was more than 0.7309, and the RMSE was less than 8.22 mg L-1. For Si-ATR, the RPD was more than 1.42, the R 2 was 0.5198, and the RMSE was less than 11.02 mg L-1. It was confirmed that all three types of ATR could be applied in the quantification of NO3 --N in water for high nitrate concentrations. However, for the quantification of low nitrate concentrations (0-1 mg L-1 NO3 --N), ZnSe-ATR and diamond-ATR acquired the same accuracy, while Si-ATR had a lower accuracy. The pretreatment of Si-ATR-based spectra using the deconvolution algorithm could improve the prediction accuracy compared to water deduction for predicting low NO3 --N. Furthermore, a Si-ATR accessory was developed using Si-ATR, which was reliable for NO3 --N concentration quantification in water with the advantage of its low cost and long durability. Totally, samples with high nitrate concentrations implied a more reliable prediction for all crystals, and comprehensively, ZnSe-ATR was recommended for sensing low nitrate concentrations; diamond-ATR was recommended for samples with strong acid or base corrosion; and for sensing relatively high nitrate concentrations, such as in natural water bodies, Si-ATR was more economical because of its low cost and relatively long use life.

Green and rapid quantitative detection of selenium in selenium-enriched kefir grain based on Fourier transform infrared spectroscopy

Rapid monitoring of total and organic selenium content in kefir grain was essential for microbial screening and selenium-enriched food development. Firstly, spectral information of selenium-enriched kefir grain was obtained using an attenuated total reflection Fourier transform infrared spectrometer. Secondly, the performance of the quantitative prediction models established by the four-variable screening method with three machine learning algorithms, respectively, was compared. For the prediction of total selenium, the competitive adaptive reweighted sampling - least squares support vector machine model performed the best, with prediction set relative coefficient (RP) and relative prediction deviation (RPD) values of 0.97 and 4.36, respectively. For the prediction of organic selenium, the IRF-LSSVM model had a RP and RPD value of 0.95 and 6.44, respectively. The proposed method achieves scientific, rapid (within 1 min) and green detection of total selenium (237.72–2330.82 μg/g) and organic selenium (102.20–1483.59 μg/g) content in selenium-enriched Kefir grain.

Rapid prediction of nucleosides content and origin traceability of Boletus bainiugan using Fourier transform near-infrared spectroscopy combined with chemometrics

Boletus bainiugan has high nutritional and economic values. As one of the potential medicinal active ingredients, nucleosides have important research significance. Porcini mushrooms fraud is frequently detected on the market, including substitute inferior into superior and lack of geographical origin’s certification. This behavior results in economic loss and health damage to consumers. Fourier transform near-infrared (FT-NIR) spectroscopy is a fast, efficient and reliable analytical tool. In the present study, the effect of source environment (climatic factors) on nucleoside content is analyzed for the first time. Then, the FT-NIR spectroscopy to study the origin traceability and content prediction of Boletus bainiugan are utilized. The results indicate that the nucleoside content is associated with precipitation and temperature. The combination of synchronous two-dimensional correlation spectroscopy (2DCOS) with residual neural networks (ResNet) model obtains the precise identification of the origin of Boletus bainiugan, with an accuracy of 100%. In the prediction models of content for uridine, guanosine, and adenosine, the optimal coefficient of determination of predictive set (R2P) is 0.901, and the optimum residual prediction deviation (RPD) is 3.178. FT-NIR spectroscopy has proven to be an environmentally friendly and non-destructive analytical tool for accurate origin traceability and content prediction.

A natural pigment-based nanosized colorimetric sensor for freshness evaluation of aquatic products

This work proposes an innovative method for monitoring the freshness of aquatic products using a nanosized colorimetric sensor (NCS). Six porous organic frameworks (POFs) were utilised to modify and sensitize four food-friendly natural pigments. Four optimal nano-pigments were selected based on their RGB-responsive interaction with NH3. Meanwhile, pigments exhibited significant changes in Vis-NIR spectra after nanosizing. Results indicated that the optimal NCS performs well in predicting aquatic products' total volatile alkaline nitrogen (TVB-N) content. While the prediction model based on image data did not benefit from nanosizing, the model constructed from spectral data demonstrated high accuracy and stability. The best PLS model achieved a prediction set correlation coefficient (Rp) of 0.9891 and a residual prediction deviation value of 6.66 by combining with variable combination population analysis–iteratively retaining informative variables (VCPA-IRIV) algorithm. Thus, the nanofabrication of POFs on pigments shows promise for developing high-precision and stable freshness prediction models.

Microwave spectroscopy in a free-space arrangement for nondestructive quality assessment of chicken eggs: Comparing different measurement modes and feature selection approaches

A free-space dielectric arrangement with X-band coaxial-to-waveguide adapters was used to non-destructively assess shell egg quality. Scattering parameters in the microwave spectral range (8–12 GHz) were measured in reflectance and transmittance modes from eggs placed in three orientations. Partial least squares regression was applied to predict egg quality indices, including air cell height (ACH), yolk coefficient (YC), thick albumen height (TAH), Haugh unit (HU), and albumen pH. Prioritizing PR_S22 spectrum in horizontal 1 orientation, several feature selection methods were employed to identify the most effective frequencies. Artificial neural networks (ANNs) were then used to develop predictive models based on influential frequencies. The competitive adaptive reweighted sampling method consistently outperformed others, yielding robust ANN models with excellent residual predictive deviation values of 4.80, 4.00, 3.27, 3.03, and 3.72 for ACH, YC, TAH, HU, and albumen pH, respectively. This study demonstrates the effectiveness of free-space dielectric arrangements in predicting egg quality.

Diversity Analysis and Comprehensive Evaluation of 101 Soybean (Glycine max L.) Germplasms Based on Sprout Quality Characteristics.

Soybean sprouts are a common culinary vegetable due to their high nutrition and tasty flavors. To select soybean varieties with excellent sprout quality, 101 soybean materials were collected from different regions of China, and eight sprout quality parameters were determined for overall evaluation. The results showed that eight sprout quality parameters achieved varying degrees of difference and correlation. Based on the principal component analysis (PCA), three principal components were extracted, with a cumulative contribution rate of 78.314%. Further, the comprehensive evaluation value (D) of soybean sprout quality was calculated by membership function analysis based on PCA, and the quality of soybean sprouts was ranked accordingly. Subsequently, a regression equation for the prediction of soybean sprout quality was established using a stepwise regression analysis, and the model showed a good prediction performance (correlation coefficient of prediction > 0.8; residual predictive deviation > 2.0). On these grounds, it was proposed that the quality of soybean sprouts could be comprehensively predicted by four parameters: hypocotyl length, edible rate, 100-seed weight, and total isoflavone content and saponin content. In conclusion, this study provides excellent varieties for soybean sprout production and new variety breeding, and it provides an important reference for the prediction of soybean sprout quality.

Determination of aflatoxin B1 in wheat using Raman spectroscopy combined with chemometrics

Aflatoxin B1 (AFB1) is carcinogenic and highly susceptible to production in wheat. In this study, the quantitative detection of contaminant AFB1 in wheat was investigated by Raman spectroscopy combined with chemometric method realization. Firstly, Savitzky–Golay smoothing (SG) and baseline calibration methods were used to perform the necessary preprocessing of the collected raw Raman spectra. Then, three variable optimization methods, i.e., competitive adaptive reweighted sampling (CARS), iteratively variable subset optimization (IVSO), and bootstrap soft shrinkage (BOSS), were applied to the preprocessed wheat Raman spectra. Finally, partial least squares regression (PLSR) models were developed to determine AFB1 in wheat samples. The results showed that all three variable optimization algorithms significantly improved the predictive performance of the models. The BOSS-PLSR model has strong generalization performance and robustness. Its prediction coefficient of determination (RP2) was 0.9927, the root mean square error of prediction (RMSEP) was 2.4260 μg/kg, and the relative prediction deviation (RPD) was 11.5250, respectively. In conclusion, the combination of Raman spectroscopy and chemometrics can realize the rapid quantitative detection of AFB1 in wheat.

Force model of robot bone grinding based on finite element analysis

Bone grinding represents a formidable challenge during open epiphyseal surgery, as grinding force may result in potential damage to bone and surrounding tissues. To address this issue, this study establishes a correlation between grinding force and bone density, as well as feed rate and rotational speed. Given the limited availability of experimental data on bone grinding, a finite element analysis (FEA) model is developed. Subsequently, the impact of bone density, feed rate and rotational speed on grinding force is determined through linear regression method. The accuracy and sensitivity of the grinding force model are then assessed, revealing a maximum prediction deviation of 4.87%. Finally, based on the established model, safe operating parameters and boundary conditions for robot bone grinding are identified.

Comparative study on the performance of fresh concrete comprising manufactured sand vs. river sand during full-scale long-distance pumping

The shift from river sand (RS) to manufactured sand (MS) as a fine aggregate in concrete is driven by environmental sustainability concerns. However, negative impacts of MS on concrete workability have raised doubts about its suitability for long-distance pumping in construction projects. This study evaluates the performance of MS concrete compared to RS concrete of the same strength grade during long-distance pumping. A full-scale coiled pipeline with measured length of 348 m was built, twelve batches of C30 and C60 concrete were prepared, and pumping tests were conducted at discharge rates of 2.5–18.9 liter per second (L/s). The workability and rheological parameters were assessed before and after pumping. The measured and predicted pumping pressure were analyzed and calculated based on rheological parameters and lubricating layer (LL) parameters. Results indicate that MS increased the yield stress and decreased the LL thickness index of C30 concrete, while it decreased the viscosity coefficient of C60 concrete. Post-pumping, the change in yield stress was similar for both MS and RS concrete, but MS reduced the plastic viscosity change in C60 concrete. MS did not significantly affect the pumping pressure loss, but influenced the pumping pressure prediction deviation by influencing the thixotropy. The classic pumping pressure prediction model developed by Kapan obviously overestimated the pumping pressure during long-distance pumping, and the deviation was more prominent in high-strength concrete pumping. MS increased the thixotropy of C30 concrete and reduced the prediction deviation of pumping pressure, but had the opposite effect on C60 concrete.

Determination of aflatoxin B1 in peanuts based on millimetre wave

This study introduces a novel method for swift detection of AFB1 in peanuts using millimetre wave technology. The research team devised a portable millimetre-wave detection device employing a double-external-difference mixing structure. The device measured millimetre-wave transmission coefficients in the 20 GHz - 40 GHz frequency range for peanut samples. Results showed that the PCA-KNN model excelled in qualitative AFB1 detection, achieving 100 % accuracy in the prediction set. In quantitative analysis, by condensing the feature variables into a 16-dimensional space, the BOSS-PSO-SVR model enhanced performance. Compared to the full transmission coefficient SVR model, the BOSS-PSO-SVR model exhibited improved coefficients of determination (RP2), reducing root mean square error of prediction (RMSEP) from 36.49 μg∙kg−1 to 19.08 μg∙kg−1, and enhancing relative prediction deviation (RPD) from 3.17 to 6.06. This study concludes that the integration of a custom miniaturized millimetre-wave device with appropriate chemometric methods facilitates rapid and accurate detection of peanut AFB1 levels.

Bargaining in the Shadow of Uncertainty

In bargaining environments with stochastic future surplus, failing to delay agreement can be inefficient when the expected future surplus is sufficiently high. Theoretically, such inefficiencies never arise under unanimity rule but can arise under majority rule. Using a laboratory experiment, we find support for these predictions, both when the unanimity rule is predicted to be more efficient and when there should be no difference between the two rules. We also find large point prediction deviations under the majority rule. We show these deviations can be explained by higher-than-predicted egalitarian sharing and a lower risk of being excluded from future agreements. (JEL C78, C92, D71, D81)