Multiple Regression Statistical Model Research Articles

Multiple regression, ANN and ANFIS models for prediction of backbreak in the open pit blasting

Backbreak is one of the undesirable effects of blasting operations causing instability in mine walls, falling down the machinery, improper fragmentation and reduction in efficiency of drilling. Backbreak can be affected by various parameters such as the rock mass properties, blasting geometry and explosive properties. In this study, the application of the artificial neural network (ANN), an adaptive neuro-fuzzy inference system (ANFIS) for prediction of backbreak, was described and compared with the traditional statistical model of multiple regression. The performance of these models was assessed through the root mean square error, correlation coefficient (R 2) and mean absolute percentage error. As a result, it was found that the constructed ANFIS exhibited a higher performance than the ANN and multiple regression for backbreak prediction.

Estimating labile particulate iron concentrations in coastal waters from remote sensing data

Owing to the difficulties inherent in measuring trace metals and the importance of iron as a limiting nutrient for biological systems, the ability to monitor particulate iron concentration remotely is desirable. This study examines the relationship between labile particulate iron, described here as weak acid leachable particulate iron or total dissolvable iron, and easily obtained bio‐optical measurements. We develop a bio‐optical proxy that can be used to estimate large‐scale patterns of labile iron concentrations in surface waters, and we extend this by including other environmental variables in a multiple linear regression statistical model. By utilizing a ratio of optical backscatter and fluorescence obtained by satellite, we identify patterns in iron concentrations confirmed by traditional shipboard sampling. This basic relationship is improved with the addition of other environmental parameters in the statistical linear regression model. The optical proxy detects known temporal and spatial trends in average surface iron concentrations in Monterey Bay. The proxy is robust in that similar performance was obtained using two independent particulate iron data sets, but it exhibits weaker correlations than the full statistical model. This proxy will be a valuable tool for oceanographers seeking to monitor iron concentrations in coastal regions and allows for better understanding of the variability of labile particulate iron in surface waters to complement direct measurement of leachable particulate or total dissolvable iron.

EFECTO DE LA EDAD AL PRIMER PARTO SOBRE RENDIMIENTOS PRODUCTIVOS Y REPRODUCTIVOS DE UN REBAÑO LECHERO DE LA REGIÓN DE LOS RÍOS

Productive and reproductive records since 1989 to 2007 from Punahue dairy farm of Austral University of Chile were used. The data were analyzed using SAS software; multiple linear regression statistical models were used to analyse the dependent variables: milk production, percentage of fat, percentage of protein calving interval. This was done to establish whether age at first calving was related to the independent variables. Least mean squares were obtained for each variable, and were compared using the t-test for paired terms. Age at first calving classified into four groups (22-26; 27-31; 32-37; 38-48 months). The mean and standard deviation of age at first calving for the studied herd were 29.28±4.83 months. Most cows (68,4%) had their first calving within the age groups 1 and 2. Age at first calving had not a significant effect on protein percentage. Least square means for the dependent variables were significantly different for milk yield, percentage of fat and calving intervaltween the groups of age at first calving.

Clinical Determinants of Radiation Dose in Percutaneous Coronary Interventional Procedures: Influence of Patient Size, Procedure Complexity, and Performing Physician

The objectives of this work were to establish the primary clinical determinants of patient radiation dose associated with percutaneous coronary interventional (PCI) and to identify opportunities for dose reduction. Use of X-ray imaging and associated radiation dose is a necessary part of PCI. Potential adverse consequences of radiation dose include skin radiation injury and predicted increase in lifetime cancer risk. Cumulative skin dose (CSD) (measured in gray [Gy] units) was selected as a measurement of patient radiation burden. Several patient-, disease-, and treatment-related variables, including 15 performing physicians, were analyzed in a multiple linear regression statistical model with cumulative skin dose CSD as the primary end point. The model results provide an estimate of the relative CSD increase (decrease) attributable to each variable. Percutaneous coronary interventions performed on 1,287 male and 540 female patients were included. Median patient age was 68.6 years, median body mass index was 29.7 kg/m(2), and median weight was 88 kg. Median CSD was 1.64 Gy per procedure for male and 1.15 Gy for female patients. Increasing body mass index, patient sex, lesion complexity, lesion location, and performing physician were significantly associated with CSD. Physicians who performed more procedures were associated with lower CSD. Several primary determinants of patient radiation dose during PCI were identified. Along with physician development of radiation-sparing methods and skills, pre-procedure dose planning is proposed to help minimize radiation dose for PCI.

The Influence of Computer Self Efficacy, Computer Experience and Interface Design to Acceptance of Electronic Banking: Empirical Study of Bank Customers in Bekasi City

The object of this research is customers on the five major banks in the city of Bekasi namely Bank Mandiri, BCA, BRI, Bank Danamon, and BNI. This study aimed to analyze the implementation of electronic banking (e-banking) with the approach of Technology Acceptance Model (TAM). The research method is a survey method with a descriptive analysis and statistical analysis. The data is processed and analyzed by multiple linear regression statistical models using statistical software. The results of this study indicate that the person's ability to use computers, and interface design does not significantly influence perceived ease of use (PEOU). Experience of computer use, relevance, security and privacy significantly influence the perceived ease of use. Relevance does not significantly influence the perception of its usefulness (PU). Interface design, and perceived ease of use significantly influence the perception of its usefulness. Perceived ease of use significantly influences the attitude of its use. Perception of usefulness does not significantly influence the attitude of its use. Attitude for its use significantly influence the real usage and acceptance of e-banking.

Investigations on surface characteristics over an Aluminium alloy in CNC turning operation

It is universally acknowledged that the performance of any machining process is usually evaluated in terms of its productivity and surface quality. Each of these controlling criteria are affected directly and in different way by the tool edge wear, fracture, chatter, surface roughness, cutting force etc., among these responses from machining operation, surface roughness is always considered as one of the most reliable measures for tool wear monitoring and breakage detection. This paper focuses on developing an effective, inexpensive surface recognition system which could be implemented in the modern manufacturing environments. Experiments have been conducted on Aluminium 6061 T6 and for measuring surface roughness based on Design of Experiments (DOE). A statistical multiple regression model has been developed for correlating the values obtained. For the purpose of correlation, DOE software has been used and ANOVA analyses have been carried out to identify the significant factors affecting surface roughness. The experimental values are analyzed in the optimizer technique which uses the Design of Experiment (DOE) principles to generate parameters depending on the surface roughness needed in the manufacturing environment. Thus in case of mass production, time for inspection could be reduced and hence the surface recognition system results in better performance.

Quantifying Wood Decay in Sydney Bluegum (Eucalyptus saligna) Trees

The evaluation of decayed wood in a tree is essential for tree risk assessment in arboriculture. It is not feasible in practice to measure the amount of wood decay in all the wood of every tree. Therefore, the capacity of a single measurement taken in cross-section to approximate the volume of decay in the wood of a whole tree is the focus of this study. The volume of wood decay in Eucalyptus saligna trees was estimated post harvest by measuring the whole tree wood density of 36 trees. Linear, logarithmic, and multiple regression statistical models were used to examine whether whole tree wood density data could be correlated with three ways of directly measuring wood decay in a single cross-section, two prior to tree felling and one post felling method. The decay estimation techniques were measured in cross-section—a Picus® Sonic Tomograph system, an IML-Resi system, and a visual method. In this study the Resi System for estimating wood decay showed a correlation with whole tree wood density in the entire tree whereas the Picus system and the visual method did not show any statistical correlation with whole tree wood density.

Model-Independent 3D Descriptors of Vertebral Cancellous Bone Architecture

High-resolution micro computed tomography has enabled measurement of bone architecture derived from 3D representations of cancellous bone. Twenty-eight vertebral bodies were obtained from four embalmed male cadavers. From 3D anaglyphs, trabecular rod thickness and length were measured and the trabecular rod Buckling index was calculated. From 3D voxel-based datasets, bone volume density, trabecular thickness, and trabecular separation were measured. Also, trabecular bone pattern factor, structural model index, connectivity density, and degree of anisotropy were calculated. Bone volume density alone explains 59% of the variability in trabecular rod Buckling index. The addition of connectivity density, trabecular separation, and structural model index, in a multiple regression statistical model, improves the explanatory power to 77%. The relationships between measures of cancellous bone architecture and a derived measure of trabecular rod strength were investigated. Morphological descriptors of cancellous bone provide a composite explanatory model of trabecular rod strength.

Assessing the solar potential of low-density urban environments in Andean cities with desert climates: The case of the city of Mendoza, in Argentina. 2nd. Part

Energy use in the built environment is globally recognized as a key issue for sustainable urban development. In temperate-cold arid regions with a generous solar resource, such as those of western Argentina, adequate design and technology can substantially reduce the energy demand for space and water heating in urban buildings. The solar potential of low-density residential urban areas in the city of Mendoza's Metropolitan Area (MMA), has been studied earlier in this research [1]. Several indicators of the solar potential were elaborated. They provide necessary information when planning and designing new urban structures or refurbishing existing ones. However, a more direct indicator, relating the available solar radiation during a heating season to the space volume to be heated, the Volumetric Insolation Factor (VIF), seems to be of most practical use as far as contributing a helpful evaluation indicator, to the above mentioned design processes. The present study follows the methodological steps used in the former research, evaluating comparatively the results of a Graphic-Computational Model and a Multiple Linear Regression Statistical Model. As in the earlier study, the good fit of both models' results clearly point at the reliability of the statistical procedure and its valuable contribution of a simplified calculation tool as its by-product.

Effects of chemical interesterification on physicochemical properties of blends of palm stearin and palm olein

Chemical interesterification is an important technological option for the production of fats targeting commercial applications. Fat blends, formulated by binary blends of palm stearin and palm olein in different ratios, were subjected to chemical interesterification. The following determinations, before and after the interesterification reactions, were done: fatty acid composition, softening point, melting point, solid fat content and consistency. For the analytical responses a multiple regression statistical model was applied. This study has shown that blending and chemical interesterifications are an effective way to modify the physical and chemical properties of palm stearin, palm olein and their blends. The mixture and chemical interesterification allowed obtaining fats with various degrees of plasticity, increasing the possibilities for the commercial use of palm stearin and palm olein.

Proximal watershed validation of a remote sensing-based streamflow estimation model

Remotely sensed time series of landscape biophysical states and radar estimates of precipitation are assessed in a statistical streamflow estimation model in four regionally proximate south-central Texas watersheds. Sandies Creek watershed (1420 km2) served as calibration of a streamflow estimation model based on 8-day composited time series of land surface (radiant) temperature (LST) and a vegetation moisture stress index (MSI) from MODIS (Moderate Resolution Imaging Spectroradiometer) satellite imagery products. These time series were hypothesized to serve as proxies for soil moisture condition antecedent to streamflow-generating precipitation events as estimated by NEXRAD (Next Generation Weather Radar) precipitation products. A linear multiple regression statistical model yielding estimation equation for observed flows at Sandies Creek was validated in 3 additional proximal watersheds of varying spatial dimension (860 km2 - 2940 km2), soils, land cover, and climatology. The validation yielded encouraging results as assessed with Pearson's r, Nash-Sutcliffe's E, and relative volume error or bias. Equivalent performance of the calibrated model was seen in the watershed mostly adjacent to Sandies Creek, the mostly similar one in terms of size, land cover, precipitation, and soils. Performance of the model diminished slightly in the more distal, and climatologically-environmentally dissimilar watersheds. The estimation model was re-calibrated for all three validation watersheds, with positive results. The contribution of LST uncertainty and NEXRAD precipitation bias to model discrepancies were evaluated along with discussion of known sources of model error.

Determinant factors of tourist attractiveness of Greek prefectures

Tourism constitutes one of the most dynamic and rapidly developing sectors of the Greek economy, making a decisive contribution to the growth of many Greek regions. The increase in tourist fl ows to all regions of the country is a vital pursuit of its regional policies, and a means for achieving regional economic development. The intense differentiation in tourist arrivals at each individual prefecture of the country constitutes an issue that is related to the more general characteristics and factors that shape the degree of tourist attractiveness of each area. The present article examines the tourism characteristics of Greek prefectures and the factors that affect related tourist fl ows and defi ne the structure of the country’s internal tourism. Furthermore, a proposal is made as to the typology of the prefectures according to their tourist resources. The determination and the analysis of factors are performed by using the multiple regression statistical model, which calculates the impact of each individual determinant on the tourist attractiveness of the prefectures. Moreover, by using hierarchical cluster analysis, two uniform territorial units of tourist attractiveness are formed, thus giving the opportunity to decision-makers to exercise a more effective tourism and regional policy. The basic conclusion that results from the article is that the presence of sandy beaches and the sea vitally contributes to the confi guration of each prefecture’s overall tourist attractiveness.

Solar induced variations of odd nitrogen: Multiple regression analysis of UARS HALOE data

A linear multiple regression statistical model is applied to estimate the solar induced component of odd nitrogen variability in the stratosphere and lower mesosphere using UARS HALOE data for 1991–2003. Consistent with earlier studies, evidence is obtained for a decadal NOx variation at the highest available latitudes (50° – 70°) that projects positively onto the solar cycle. This variation, which is most statistically significant in the Southern Hemisphere, also correlates positively with the auroral Ap index. It is therefore probably caused by downward transport during the polar night of thermospheric and mesospheric odd nitrogen. In addition, at low latitudes near and above the stratopause, evidence is obtained for an inverse solar cycle NOx variation. It is suggested that this low‐latitude response may be caused primarily by increased photolysis of NO under solar maximum conditions. Throughout most of the rest of the stratosphere, no statistically significant response is obtained.

Changes in frost days in simulations of twentyfirst century climate

Global coupled climate model simulations of twentieth and twentyfirst century climate are analyzed for changes in frost days (defined as nighttime minima less than freezing). The model simulations agree with the observed pattern for late twentieth century of a greater decrease of frost days in the west and southwest USA compared to the rest of the country, and almost no change in frost days in fall compared to relatively larger decreases in spring. Associated with general increases of nighttime minimum temperatures, in the future climate with increased greenhouse gases (GHGs) the number of frost days is fewer almost everywhere, but there are greatest decreases over the western parts of the continents. The numbers of frost days are most consistently related to sea level pressure, with more frost days occurring when high pressure dominates on the monthly time scale in association with clearer skies and lower nighttime minimum temperatures. Spatial patterns of relative changes of frost days are indicative of regional scale atmospheric circulation changes that affect nighttime minimum temperatures. Increases of soil moisture and clouds also contribute, but play secondary roles. The linkages among soil moisture, clouds, sea level pressure, and diurnal temperature range are quantified by a statistical multiple regression model. Coefficients for present and future climate are similar among the predictors, indicating physical processes that affect frost days in present and future climates do not appreciably change. Only the intercept changes in association with the significant warming of the mean climate state. This study highlights the fact that, though there is a general decrease in the number of frost days with global warming, the processes that affect the pattern of those changes, and thus the regional changes of frost days, are influenced by several interrelated physical processes, with changes in regional atmospheric circulation generally being most important.

Investigating tendon fascicle structure-function relationships in a transgenic-age mouse model using multiple regression models.

Proper replacement or repair of damaged tendons or ligaments requires functionally engineered tissue that mimics their native mechanical properties. While tendon structure-function relationships are generally assumed, there exists little quantitative evidence of the roles of distinct tendon components in tendon function. Previous work has used linear correlations to assess the independent, univariate effects of one structural or one biochemical variable on mechanics. The current study's objective was to simultaneously and rigorously evaluate the relative contributions of seven different structural and compositional variables in predicting tissue mechanical properties through the use of multiple regression statistical models. Structural, biochemical, and mechanical analysis were all performed on tail tendon fascicles from different groups of transgenic mice, which provide a reproducible, noninvasive, in vivo model of changes in tendon structure and composition. Interestingly, glycosaminoglycan (GAG) content was observed to be the strongest predictor of mechanical properties. GAG content was also well correlated with collagen content and mean collagen fibril diameter. Collagen fibril area fraction was a significant predictor only of material properties. Therefore, in a large multivariate model, GAG content was the largest predictor of mechanical properties, perhaps both through direct influence and indirectly through its correlation with collagen content and fibril structure.

A statistical model to compare road mortality in OECD countries

The objective of this paper is to compare safety levels and trends in OECD countries from 1980 to 1994 with the help of a statistical model and to launch international discussion and further research about international comparisons. Between 1980 and 1994, the annual number of fatalities decreased drastically in all the selected countries except Japan (+12%), Greece (+56%) and ex-East Germany (+50%). The highest decreases were observed in ex-West Germany (−48%), Switzerland (−44%), Australia (−40%), and UK (−39%). In France, the decrease in fatalities over the same period reached 34%. The fatality rate, an indicator of risk, decreased in the selected countries from 1980 to 1994 except in the east-European countries during the motorization boom in the late 1980s. As fatality rates are not sufficient for international comparisons, a statistical multiple regression model is set up to compare road safety levels in 21 OECD countries over 15 years. Data were collected from IRTAD 1 IRTAD Web site, 1998. http://www.oecd.org//dsti/sti/transpor/road/stats/IRTAD 1 (International Road Traffic and Accident Database) and other OECD statistical sources. The number of fatalities is explained by seven exogenous (to road safety) variables. The model, pooling cross-sectional and time series data, supplies estimates of elasticity to the fatalities for each variable: 0.96 for the population; 0.28 for the vehicle fleet per capita; −0.16 for the percentage of buses and coaches in the motorised vehicle fleet; 0.83 for the percentage of youngsters in the population; −0.41 for the percentage of urban population; 0.39 for alcohol consumption per capita; and 0.39 for the percentage of employed people. The model also supplies a rough estimate of the safety performance of a country: the regression residuals are supposed to contain the effects of essentially endogenous and unobserved variables, independent to the exogenous variables. These endogenous variables are safety performance variables (safety actions, traffic safety policy, network improvements and social acceptance). A new indicator, better than the mortality rate, is then set upon the residuals. Mean estimates of this indicator for the years 1980–1982 and the years 1992–1994 rank the countries in the beginning and at the end of the study period. Countries showing the best ranks (and thus the best performance) in 1980 and 1994 are Sweden, the Netherlands and Norway. The UK and Switzerland reach the top 5 in 1994. Greece, Belgium, Portugal and Spain are the last countries in the classification along with, surprisingly, the USA. France was ranked 18th in 1980 and 15th in 1994 but is ranked amongst the five countries that most improved from 1980 to 1994. This model remains non definitive because it is not able to distinguish between safety performance and unobserved exogenous variables although these exogenous variables could explain more about the differences in levels and trends between the countries. More complex models, particularly highly sophisticated models regarding the number of fatalities with breakdowns by road users or road classes would be needed to give a precise and profound ranking of safety levels and safety improvements between countries.

QBO and Pinatubo signals in the mass flux at 100hPa during the period from 1986 to 1995

The residual mean meridional circulation is calculated by the transformed Eulerian-mean (TEM) momentum and continuity equations for 10 years from December 1985 to November 1995. The tropical upward mass flux at 100 hPa is estimated from the extratropical downward mass fluxes in the southern and northern hemispheres. The multiple regression statistical model is used to obtain quantitatively the signal of the quasi-biennial oscillation (QBO) and the trend. The results show that the tropical upward mass flux is enhanced during QBO easterly phase and reduced during QBO westerly phase. The signal of Mount Pinatubo volcanic eruption in June 1991 is detected. An abrupt increase of the mass flux is shown in the latter half of 1991, and the increase ceases at the end of 1993, about 2.5 years after the eruption.

The impact of climate variability and change on crop yield in Bulgaria

During the recent decade, the problem of climate variability and change, due to natural processes as well as factors of anthropogenetic origin, has come to the forefront of scientific problems. The objective of this study was to investigate climate variability in Bulgaria during the 20th century and to determine the overall impact on agriculture. There was no significant change in the mean annual air temperature. In general, there was a decrease in total precipitation amount during the warm-half of the year, starting at the end of the 1970s. Statistical multiple regression models, describing the relationship between crop yield, precipitation, and air temperature were also developed. Several transient climate change scenarios, using global climate model (GCM) outputs, were created. The Decision Support System for Agrotechnology Transfer (DSSAT) Version 3.5 was used to assess the influence of projected climate change on grain yield of maize and winter wheat in Bulgaria. Under a current level of CO 2 (330 ppm), the GCM scenarios projected a decrease in yield of winter wheat and especially maize, caused by a shorter crop growing season due to higher temperatures and a precipitation deficit. When the direct effects of CO 2 were included in the study, all GCM scenarios resulted in an increase in winter wheat yield. Adaptation measures to mitigate the potential impact of climate change on maize crop production in Bulgaria included possible changes in sowing date and hybrid selection.

QBO and Pinatubo signals in the mass flux at 100 hPa and stratospheric circulation

The residual mean meridional circulation is calculated by the transformed Eulerian‐mean (TEM) momentum and continuity equations for the period from December 1988 to November 1995. Our method does not rely on radiative calculations. The tropical upward mass flux at 100 hPa is estimated from the extratropical downward mass fluxes in the southern and northern hemispheres. The multiple regression statistical model is used to obtain quantitatively the signal of the quasi‐biennial oscillation (QBO) and the trend. The results show that the tropical upward mass flux is enhanced during QBO easterly phase and reduced during QBO westerly phase. As a residual, the signal of Mount Pinatubo volcanic eruption in June 1991 is detected. This reflects the aerosol effects to the residual circulation in the lower stratosphere after the eruption. An abrupt increase of the mass flux is shown in the latter half of 1991, and the increase ceases at the end of 1993, about 2.5 years after the eruption.

Approximate separation of volcanic and 11‐year signals in the SBUV‐SBUV/2 total ozone record over the 1979‐1995 Period

The combined Nimbus 7 SBUV, NOAA 11, and preliminary NOAA 9 SBUV/2 monthly zonal mean total ozone data set, extending from January 1979 to December 1995, is analyzed with a multiple regression statistical model that includes a term to describe the direct effect of volcanic aerosols on total ozone. Outside of polar regions, which are not included in this analysis, the volcanic regression coefficients in the northern hemisphere are negative (consistent with heterogeneous chemical losses of ozone on volcanic sulfate aerosols) and reach their maximum values between 40°N–60°N latitude. Although inclusion of the aerosol term in the statistical model introduces some additional uncertainty to the derived solar response, statistically significant values of the solar coefficients are found in northern and southern subtropical regions throughout the year, with maximum amplitudes of ∼2.5% at 30° latitude in both hemispheres during winter. We conclude that the 11‐year signal in the SBUV‐SBUV/2 total ozone record is approximately separable from volcanic aerosol effects.