Traditional Multiple Regression Analysis Research Articles

Factors driving antimony accumulation in soil-pakchoi and wheat agroecosystems: Insights and predictive models

The accumulation of antimony (Sb) in plants and its potential effects on human health are of increasing concern. Nevertheless, only a few countries or regions have established soil Sb thresholds for agricultural purposes, and soil properties have not been taken into account. This study investigated the accumulation of Sb in the edible parts of pakchoi and wheat grain by adding exogenous Sb to 21 soils with varying properties. The results revealed a positive correlation between bioavailable Sb (Sbava, extracted by 0.1 M K2HPO4) in soil and Sb in the edible parts of pakchoi (R2 = 0.77, p < 0.05) and wheat grain (R2 = 0.54, p < 0.05). Both machine learning and traditional multiple regression analysis indicated Sbava was the most critical feature and the main soil properties that contributed to Sb uptake by pakchoi and wheat were CaCO3 and clay, respectively. The advisory food limits for Sb in pakchoi and wheat were estimated based on health risk assessment, and used to derive soil thresholds for safe pakchoi and wheat production based on Sbtot and Sbava, respectively. These findings hold potential for predicting Sb uptake by crops with different soil properties and informing safe production management strategies.

Sustainable Digital Marketing Model of Geoenergy Resources under Carbon Neutrality Target

Geoenergy resources are a new type of clean energy. Carbon neutralization and carbon peaking require significant system reform in the field of energy supply. As a clean, low-carbon, stable and continuous non carbon-based energy, geothermal energy can provide an important guarantee for achieving this goal. Different from the direct way of obtaining energy, ground energy indirectly obtains heat energy from shallow soil and surface water. The vigorous development of geoenergy resources under China’s carbon neutrality goal plays an important role in energy conservation and emission reduction. However, the current carbon trading market is not understood by the public. Therefore, this paper aims to analyze the impact of geoenergy resources on promoting sustainable digital marketing models. Every country around the world is working hard to meet its carbon neutrality goal. This paper analyzed the economic goal of carbon neutrality by analyzing the principle of the carbon trading market. For this reason, this paper designed a carbon trading price prediction algorithm based on the BP neural network (BPNN), which can predict prices in the carbon trading market in order to promote the accurate push of the digital marketing model and finally drive ground energy resources to promote a sustainable digital marketing model. The experimental results of this paper have proved that the price error rate of the BPNN carbon trading price prediction algorithm designed in this paper was within 10%, which was about 20% smaller than the traditional multiple regression analysis algorithm. This proved that the algorithm in this paper has a good performance and can provide accurate information to allow the digital marketing model to achieve sustainable digital marketing.

Correlation between air pollution and prevalence of conjunctivitis in South Korea using analysis of public big data

This study investigated how changes in weather factors affect the prevalence of conjunctivitis using public big data in South Korea. A total of 1,428 public big data entries from January 2013 to December 2019 were collected. Disease data and basic climate/air pollutant concentration records were collected from nationally provided big data. Meteorological factors affecting eye diseases were identified using multiple linear regression and machine learning analysis methods such as extreme gradient boosting (XGBoost), decision tree, and random forest. The prediction model with the best performance was XGBoost (1.180), followed by multiple regression (1.195), random forest (1.206), and decision tree (1.544) when using root mean square error (RMSE) values. With the XGBoost model, province was the most important variable (0.352), followed by month (0.289) and carbon monoxide exposure (0.133). Other air pollutants including sulfur dioxide, PM10, nitrogen dioxides, and ozone showed low associations with conjunctivitis. We identified factors associated with conjunctivitis using traditional multiple regression analysis and machine learning techniques. Regional factors were important for the prevalence of conjunctivitis as well as the atmosphere and air quality factors.

Research on predictive control of students’ performance in PE classes based on the mathematical model of multiple linear regression equation

Abstract Aiming to solve the problems in the traditional multiple regression analysis model for predicting college sports performance based on the principles of econometrics, a predictive model that combines genetic algorithm (GA), college sports performance evaluation and regression analysis is proposed. GA is used to conduct dynamic and supervised optimisation evaluation of college sports performance; on this basis, combined with regression analysis and GA's global optimisation capabilities, a complex nonlinear relationship between student sports performance and influencing factors is established; the student's performance is calculated based on the college sports performance. The results show that the method has high prediction accuracy and good stability.

Water Quality Analysis of Drinking Water Resource Lake Sapanca and Suggestions for the Solution of the Pollution Problem in the Context of Sustainable Environment Approach

Lake Sapanca is the drinking water source of the Sakarya province of Turkey. Intensive urbanization in the region is the main obstacle to implementing appropriate physical planning and measures to adapt to rapid change. The monitoring of the water quality parameters in the planning and management of the lakes is significant. The Artificial Neural Network (ANN), a mathematical representation of the human brain’s functioning, was employed to estimate the Lake’s Dissolved Oxygen (DO) concentration. pH, Magnesium (Mg), Temperature (Temp), Chemical Oxygen Demand (COD), Orthophosphate (o-PO4), Nitrite Nitrogen (NO2-N), and Nitrate Nitrogen (NO3-N) were used as independent parameters. The successful ANN model gives better results compared to the traditional multiple linear regression (MLR) analysis. The developed model can be used for forecast purposes to complete the missing data in the future and support the decision process for pollution reduction through sustainable environmental management. The eutrophication threat for Lake Sapanca has been revealed. The main objective is to create the scientific infrastructure that will draw attention to the rapid urbanization problem with ANN and eutrophication models’ outputs. It has been understood that the protection of the water budget of Lake Sapanca is the primary solution method in terms of ecological sustainability to eliminate the existing pollution.

Application of principal component analysis for predicting body weight of Ethiopian indigenous chicken populations.

Under smallholder management conditions, where weighing scale is not readily available, body weight (BW) can be predicted from morphometric measurements using multiple regression. However, the statistical interpretation of the regression parameters estimated by least squares is difficult due to multicollinearity problems. Thus, principal component analysis (PCA) was used to predict BW from body measurements, including body length (BL), chest circumference (CC), shank length (SL), and shank circumference (SC) of Ethiopian indigenous chicken populations reared by smallholder farmers. The effectiveness of this technique was also compared with the traditional multiple regression analysis (MRA). Measurements were taken from 134 male and 487 female chickens. The whole dataset was partitioned into two portions, namely, training and testing datasets, for model comparison and validation purposes. The training dataset, which consisted of 75% of the dataset, was used to develop the model, and the testing dataset, which consisted of 25% of the dataset, was used to validate the model. The PCA results revealed that the variables for body measurements were represented by PC1 and PC2 in male birds and PC1, PC2, and PC3 in female birds. Regression models developed using scores derived from these PCs explained 88 and 69% of total variation in BW in male and female birds, respectively. Compared with traditional MRA, regression models generated using the PCA procedure were more accurate in predicting BW. Thus, the results of the present study could not only be used for predicting BW of Ethiopian indigenous chickens but also in their genetic selection programs.

The impact of regional industrial structure differences on carbon emission differences in China: An evolutionary perspective

All industries produce different carbon emissions during their production process. Previous studies use multivariate regression or panel models to explore the impact of the industrial structure on regional carbon emissions differences based on time series data or panel data from the perspective of one-dimensional continuous variables. However, there are also regional differences in the factors affecting regional carbon emission differences, such as the economic development level, the industrial structure, and urbanization. The regional carbon emission difference matrix and the influencing factor difference matrix are relational variables, and it is difficult for traditional multiple regression analysis to study the relationship between relational variables. Thus, this paper proposes the perspective of two-dimensional difference relations for studying the influencing factors of regional carbon emission differences based on Quadratic Assignment Procedure model. Energy consumption and the value added by industrial processes in various industries from 30 regions in China are taken as the sample. The results of this study are as follows: carbon emissions are increasing and the differences in carbon emissions between regions are also increasing. There are obvious regional differences in the industrial structure, economic growth, population and urbanization in China. The influence and magnitude of the impact of regional industrial structure differences on regional carbon emission differences vary from year to year. Interregional differences in industry, construction and transportation and warehousing industries had a significantly positive impact on interregional differences in carbon emissions from 2007 to 2016. By analyzing the impact of the influence mechanism of regional industrial structure differences on carbon emissions, the findings of this study can identify some suggestions for regionally balanced emission reduction policies.

Decision Tree and Boosting Techniques in Artificial Intelligence Based Automated Valuation Models (AI-AVM)

This paper develops an artificial intelligence-based automated valuation model (AI-AVM) using the decision tree and the boosting techniques to predict residential property prices in Singapore. We use more than 300,000 property transaction data from Singapore’s private residential property market for the period from 1995 to 2017 for the training of the AI-AVM models. The two tree-based AI-AVM models show superior performance over the traditional multiple regression analysis (MRA) model when predicting the property prices. We also extend the application of the AI-AVM to more homogenous public housing prices in Singapore, and the predictive performance remains robust. The boosting AI-AVM models that allow for inter-dependence structure in the decision trees is the best model that explains more than 88% of the variance in both private and public housing prices and keep the prediction errors to less than 6% for HDB and 9% for the private market. When subject the AI-AVM to the out-of-sample forecasting using the 2017-2019 testing property sale samples, the prediction errors remain within a narrow range of between 5% and 9%.

The confidence affects willingness to use pre-exposure prophylaxis among MSM, based on Path analysis

Objective: To analyze the influence of the confidence to pre-exposure prophylaxis (PrEP) on the willingness to use PrEP in men who have sex with men (MSM) by using the path analysis model based on structural equation model approach. Methods: A total of 550 MSM were recruited in Urumqi by snowball method and a questionnaire survey was conducted among them. According to the professional knowledge and analysis results of the confidence and willingness to use PrEP, structural equation model (SEM) analysis method was used to construct a path analysis model. Results: A total of 513 MSM participated in the survey. The modified path equation model was well fitted, with the modified fitting index as: GFI=0.993, RMSEA<0.001, and AGFI=0.984. The confidence to PrEP had direct influence on the willingness; the degree of influence from sex partners, the attitude of sex partner to PrEP and the positive emotions not only had direct effects on willingness of PrEP use, but also had indirect effects on willingness of PrEP use by affecting the confidence to it; the role in sexual behavior, AIDS severity, HIV prevention behavior had direct effects on willingness of PrEP use. The proportion of HIV infection in the population had no direct effects on willingness of PrEP use, but had indirect effects on willingness of PrEP use by affecting the confidence to it. Conclusions: The confidence to PrEP had influence on willingness of PrEP use in MSM, therefore targeted activities can be conducted to improve the confidence and willingness of MSM in taking the PrEP and reducing the risk of HIV infection in MSM. Compared with the traditional multiple regression analysis, the path analysis using the structural equation model could better reveal the mediating effect between the independent variables and dependent variables.

FsQCA versus regression: The context of customer engagement

The purpose of this study is to address the strengths and weaknesses of the Fuzzy set Qualitative Comparative Analysis (fsQCA) method by contrasting it with traditional multiple regression analysis. To achieve our objective, we utilize the context of customer engagement and explore how both methodological approaches can explain the factors that lead customers to engage with brands/firms. We test the association between customer engagement and its antecedents by using both methods, fsQCA and multiple regression analysis (MRA), and compare the findings. MRA results show a significant relationship between customer engagement and perceived interactivity, perceived identification, perceived value, and brand involvement, a negative relationship with perceived fairness, while the relationship with customer satisfaction and perceived value is not statistically significant. On the other hand, fsQCA revealed eight different combinations of these antecedents that lead to high levels of customer engagement. Although multiple regression analysis (MRA) certainly has its own merit, our findings indicate that this methodological approach does not capture the complexities extant in real life. This method's strength of providing generalizable results is also its weakness as relationships cannot always be generalized as shown in our study. As such, at a minimum, fsQCA should be used as a complementary approach when authors employ MRA. We suggest ‘at a minimum’ because fsQCA possesses sufficient explanatory power to be employed by itself to offer a rich perspective on the relationships of interest. We also make several noteworthy contributions to the customer engagement and service marketing literatures.

Predicting deformational properties of Indian coal: Soft computing and regression analysis approach

Deformational properties, such as, Young’s modulus and Poisson’s ratio are considered to be among the most critical parameters of coal measure rocks for coal bed methane (CBM) extraction, CO2 sequestration, and mining activities. The inherent heterogeneity of these rocks from basin to basin makes the accurate estimation of deformational properties very complex. Here, these properties were evaluated using artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS), and compared with the traditional multiple regression analysis (MRA). Multiple index geomechanical parameters, such as, compressive strength, tensile strength, shear strength, and P-wave velocity have been used as the inputs to build the prediction models. In total 69 data sets, developed from laboratory measurements were used for the analysis. First 57 datasets were used to train the soft computing models, and the rest 12 were used to measure the accuracy and validity. Performance of the each of the statistical and soft computing models were evaluated using correlation coefficient (R2), root mean square error (RMSE), variables accounted for (VAF), and mean absolute percentage error (MAPE). Results show that ANFIS model gives the best prediction model for both the Young’s modulus and Poisson’s ratio. Therefore, the present study shows that it is possible to build accurate and generalized soft computing model than can incorporate heterogeneity and predict deformational properties of coal measure rocks very well without resorting to time consuming test procedures.

Development of Equations for Meander Movement Velocities by SCE Optimization Algorithm based on Experimental Study

Meandering rivers have a dynamic planform, which means that rivers move in radial direction with variable velocities. The movement velocities can be estimated from historical maps of specific river or using mathematical approach. There is not explicit method or suitable algebraic equation for estimation of river meander movement velocities. In this study, experiments were carried out in a laboratory flume, to develop novel equations to calculate meandering river movement velocities. We designated these empirical equations as the river resistance equations. The experiments were conducted with non-uniform, nonchosive sediment in a re-circulating flume with changeable slopes. The influence of bankfull discharge, bed slope and channel width on the meander movement velocities were discussed. Based on the experimental data, three novel empirical equations for prediction of Wx and Wexp and Wm were proposed. Moreover, the shuffled complex evolution (SCE) optimization algorithm was used to estimate the unknown parameters of the empirical equations. To validate SCE method, the result was compared with traditional multiple regression analysis. This comparison indicated that the SCE optimization algorithm outperformed the multiple regression analysis method in the empirical equations calibration.

Resampling Techniques for Real Estate Appraisals: Testing the Bootstrap Approach

Applied to real estate markets analysis, the resampling methods aim to contribute to the knowledge growth of real estate market dynamics, overcoming the issues related to the data scarcity and operational limits of traditional statistical theory. Among resampling methods, the Bootstrap technique appears to be the most suitable for the interpretation of real estate phenomena. In this study, for residential properties located in Cosenza (Calabria Region, Italy), a Bootstrap approach has been used in order to determine the marginal prices of the real estate characteristics detected, comparing the results with those obtainable with a traditional Multiple Regression Analysis.

Real Estate Appraisals with Bayesian Approach and Markov Chain Hybrid Monte Carlo Method: An Application to a Central Urban Area of Naples

This paper experiments an artificial neural networks model with Bayesian approach on a small real estate sample. The output distribution has been calculated operating a numerical integration on the weights space with the Markov Chain Hybrid Monte Carlo Method (MCHMCM). On the same real estate sample, MCHMCM has been compared with a neural networks model (NNs), traditional multiple regression analysis (MRA) and the Penalized Spline Semiparametric Method (PSSM). All four methods have been developed for testing the forecasting capacity and reliability of MCHMCM in the real estate field. The Markov Chain Hybrid Monte Carlo Method has proved to be the best model with an absolute average percentage error of 6.61%.

An adaptive algorithm for single-trial P300 evoked potential extraction based on dynamic feature library

Objective The single-trial extraction method of evoked potential has been one of the problems in EEG information processing field. According to the characteristics of somatosensory evoked electroencephalogram (EEG) with low signal-to-noise ratio and large parameter variation between trials, a novel single-trial extraction method for evoked potentials was proposed. This method aims to further improve the accuracy and characteristics of the single-trial extraction algorithm, preserve more dynamic characteristics between trials, and improve the estimation accuracy. Methods Based on wavelet filtering and multiple linear analysis, a new single-trial extraction method for EEG P300 parameters was proposed by applying the adaptive dynamic feature library. Four groups of wavelet filtered evoked EEG data were randomly selected, and used to build the feature library using overlapping average method and principal component analysis. For the single-trial extracted EEG data, the component with the highest correlation coefficient related with the current data was selected as the independent variable from the feature library, and the relevant multiple linear regression analysis was conducted. The single-trial evoked potential signal was reconstructed by the regression analysis results, in which the key features such as latency and amplitude were automatically extracted. Results Compared with the benchmark values determined by experts, the proposed algorithm can obtain more accurate estimation values of latency and amplitude in P300 components. The average difference of latency and amplitude by the proposed algorithm is (11.16±8.60) ms and (1.40±1.34) μV, respectively. These two values obtained by the proposed algorithm are much closer to that obtained by the commonly used overlapping average method of (23.26 ± 25.76) ms and (2.52 ± 2.50) μV, respectively. These results show that the proposed algorithm has significant advantages comparing with the traditional multiple linear regression analysis algorithm. Conclusions The dynamic updating principal component sample library of EEG data was applied to wavelet filtering and multiple linear regression, thus the dynamic characteristics were effectively preserved, and the accuracy of parameter estimation was improved. Key words: Multiple linear regression with dispersion terms; Evoked electroencephalogram; P300; Principal component analysis

Propensity Score Matching in Accounting Research

ABSTRACT Propensity score matching (PSM) has become a popular technique for estimating average treatment effects (ATEs) in accounting research. In this study, we discuss the usefulness and limitations of PSM relative to more traditional multiple regression (MR) analysis. We discuss several PSM design choices and review the use of PSM in 86 articles in leading accounting journals from 2008–2014. We document a significant increase in the use of PSM from zero studies in 2008 to 26 studies in 2014. However, studies often oversell the capabilities of PSM, fail to disclose important design choices, and/or implement PSM in a theoretically inconsistent manner. We then empirically illustrate complications associated with PSM in three accounting research settings. We first demonstrate that when the treatment is not binary, PSM tends to confine analyses to a subsample of observations where the effect size is likely to be smallest. We also show that seemingly innocuous design choices greatly influence sample composition and estimates of the ATE. We conclude with suggestions for future research considering the use of matching methods. Data Availability: All data used are available from sources cited in the text.

Examining the nature and directionality of the relationship between children's anxiety symptoms and the immunoendocrine systems

Background: Preterm birth is strongly associated with mortality, morbidity and lifetime difficulties including neurodevelopmental deficits. Immaturity of the lungs at birth is a major underlying cause, thus antenatal corticosteroid therapy (ACT) is administered to pregnant women in cases of threatened preterm birth to accelerate fetal lung development in hope of lessening infant mortality and morbidity. However, ACT has become controversial due to potential side-effects of exposure to excessive corticoids for the fetus. Objective: To study the associations between antenatal exposure to corticosteroids and infant size at birth. Design, setting, and participants: Population-based cohort study of women with live singleton births in Finland from 2006 through 2010. Maternal, pregnancy, and obstetric data came from the nationwide Finnish Medical Birth Register. Methods: Two analytic strategies were compared. First, traditional multiple regression analysis was employed to test the associationsadjusted forpotential confounders. Second,propensity scorematching (PSM)was used to take into account pre-treatment factors that could influence treatment allocation and outcome. Because PSM analysis closely matches participants on treatment characteristics results are akin to a randomized control trial although observational data are used. Results: We obtained data for 278508 women and their liveborn singletons (4887 treated and 273621 untreated). Results of adjusted multiple regression analyses and PSM converged. There were no differences by treatment in birth size for infants born very early (gestational age 24-29 weeks). Preterm infants born between 30 and 37 weeks exposed to ACT were smaller than untreated infants of the samegestational age in all respects: 518g lighter birth weight, 2.68 cm shorter birth length, 0.12 lower ponderal index, and 1.77 cm smaller head circumference, according to PSM.Deficits were also statistically significant among term born infants (gestational weeks 38-41) though differences were of lower magnitude. Post-term infants were of similar size regardless of ACT exposure. Conclusions and relevance: We were able to capitalize on the clinical practice of ACT during pregnancy to test whether synthetic glucocorticoids impact on the growth potential of the developing fetus. The results showed that infantswere smaller at birth and that head circumference, indicative of brain volume,was compromised. These results lend support to the hypothesis that prenatal exposure to excessive cortisol affects infant size and neurodevelopment.

气象因子对江苏省水稻单产的影响分析

PDF HTML阅读 XML下载 导出引用 引用提醒 气象因子对江苏省水稻单产的影响 DOI: 10.5846/stxb201309212315 作者: 作者单位: 南京师范大学地理科学学院 南京,210046 作者简介: 通讯作者: 中图分类号: F307. 11 基金项目: 国家自然科学基金项目(41271189, 40971231, 40971105); 国家863项目(2007AA12Z207); 江苏省高校优势学科建设工程项目 Meteorological effects on rice yields in Jiangsu Province Author: Affiliation: College of Geographical Science,Nanjing Normal University Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:全球气候变暖作为一个不争的客观事实,不可避免地对农业生产产生影响。针对传统多元线性回归分析方法,不能直接分析气象因子与水稻气象单产时序关系,根据1978-2010年间江苏省水稻单产数据和同期气象时序数据,研究了水稻单产的影响因素,提取了水稻气象单产;利用连续小波变换方法研究了水稻气象单产、水稻营养生长与生殖生长期间的日照时数、降水和气温等气象因子的时序变化特征;利用交叉小波和相干小波变换方法,研究了水稻营养生长与生殖生长期间气象因子与水稻气象单产间的相互影响关系。结果显示:(1)近33 a间,江苏省水稻气象单产占实际单产的比重逐渐减小,水稻生产抵御气象灾害能力逐渐增强。(2)水稻气象单产与日照时数、降水量和气温等气象因子有几乎一致的特征周期。(3)在水稻分蘖期、孕穗期与开花结实期,气象单产与日照时数、降水量和气温间的相位差关系较为复杂。水稻分蘖期日照时数的增多有利于水稻单产的增加,降水的增多导致水稻单产的下降。水稻开花结实期日照时数的增强、昼夜温差的变大有利于水稻单产的增加,夜间最低气温的上升会导致水稻单产的下降。为了应对全球气候变暖,需要进一步改变水稻种植方式,加强土地利用监管,积极开展农村土地综合整治,加强高标准基本农田建设,加大农田水利设施建设,调整作物播种期,加强气象灾害应对防范体系建设,更好地发挥生物技术在适应气候变化中的作用。 Abstract:Global warming impacts agricultural grain security production in China. We investigated the time series relationship between rice yields and meteorological factors, which are not directly analyzed using traditional multiple linear regression analysis. We obtained rice yield data from the Statistical Yearbook of Jiangsu Province and meteorological data from the China Meteorological Data Sharing Service Network for the period 1978 to 2010. We applied continuous wavelet transform, cross-wavelet and coherent-wavelet methods to analyze changes in rice yields related to meteorological factors such as sunshine hours, precipitation and temperature during rice vegetative and reproductive growth stages. Our results show that the proportion of rice yields affected by meteorological factors in Jiangsu Province decreased from 1978 to 2010 because the ability to withstand weather disasters improved. The continuous wavelet transform period characteristics of rice meteorological yields and meteorological factors such as sunshine hours, precipitation, temperature, the highest temperature, the lowest temperature,and day/night temperature difference are almost similar. Our cross-wavelet and coherent-wavelet transform analyses indicate that the phase difference relationship between rice yield and meteorological factors is complex during seedling tillering, booting development, and blossom fruiting stages. During the seedling tillering stage, an increase in sunshine hours increases rice yields, and an increase in precipitation decreases rice yields. During the blossom fruiting stage, an increasing amount of sunshine hours and large day/night temperature differences increase rice yields; small day/night temperature differences decrease rice yields. To mitigate the effects of global warming, rice cultivation procedures need to change. Land use supervision and management as well as rural land consolidation and high-standard basic farmland construction should be strengthened, farmland water conservation facilities need to be improved, and crop sowing times should be adjusted. Technological devices to predict extreme weather disasters should continue to improve and new biotechnology techniques should be implemented to ensure that rice yields are high and are not severely affected by climate change. Cross wavelet analysis methods can accurately analyze the relationship between meteorological factors and the rice yields; such analysis also helps to predict rice yield outcomes from current meteorological data. Our results enable officials in government agricultural departments to make informed decisions while implementing new agricultural policies related to the storage and harvesting of rice. 参考文献 相似文献 引证文献

Prediction accuracy in mass appraisal: a comparison of modern approaches

The advancement of computational software within the last decade has facilitated enhanced uptake of mass appraisal methodologies by the valuation and prediction accuracy in computer-assisted mass appraisal community for price modelling, estimation and tribunal defence. Applying a sample of 2694 residential properties, this paper assesses and analyses a number of geostatistical approaches relative to an artificial neural network (ANN) model and the traditional linear hedonic pricing model for mass appraisal valuation accuracy and price estimation purposes. The findings demonstrate that the geostatistical localised regression approach is superior in terms of model explanation, reliability and accuracy. ANNs can be shown to perform very well in terms of predictive power, and therefore valuation accuracy, outperforming the traditional multiple regression analysis (MRA) and approaching the performance of spatially weighted regression approaches. However, ANNs retain a ‘black box’ architecture that limits their usefulness to practitioners in the field. In relation to cost-effectiveness and user-friendly applicability for the valuation community, the MRA approach outperforms the ‘black box’ nature of the ANN technique, with the geographically weighted regression approach providing the best balance of outright performance and transparency of methodology. It is this spatially weighted approach utilising absolute location which appears to represent the way forward in developing the practice of mass appraisal.

Applying Back-Propagation Neural Network for Estimating the Slump of Concrete

This paper proposes the back-propagation neural network (BPN) and applies it to estimate the slump of high-performance concrete (HPC). It is known that HPC is a highly complex material whose behaviour is difficult to model, especially for slump. To estimate the slump, it is a nonlinear function of the content of all concrete ingredients, including cement, fly ash, blast furnace slag, water, superplasticizer, and coarse and fine aggregate. Therefore, slump estimation is set as a function of the content of these seven concrete ingredients and additional four important ratios. The results show that BPN obtains a more accurate mathematical equation through learning procedures which outperforms the traditional multiple linear regression analysis (RA), with lower estimating errors for predicting the HPC slump.