Página 1 dos resultados de 259 itens digitais encontrados em 0.013 segundos

Predicting the carcass composition of lambs by a simultaneous equations model

Cadavez, Vasco
Fonte: The National Institute of Research and Development for Biology and Animal Nutrition Publicador: The National Institute of Research and Development for Biology and Animal Nutrition
Tipo: Conferência ou Objeto de Conferência
Português
Relevância na Pesquisa
26.28%
The objective of this study was to develop models to predict lamb carcass composition by simultaneous equations model (SEM), and to compare t he efficiency of the ordinary least squares (OLS), weight least squares (WLS), and seemingly unrelated regressions (SUR) estimators. Forty male lambs, 22 of Churro Galego Bragançano Portuguese local breed and 18 of Suffolk breed were used. Lambs were slaughtered and carcasses were weighed approximately 30 min after slaughter in order to obtain hot carcass weight (HCW). After cooling at 4°C for 24-h, the subcutaneous fat thickness measurement (C3) was taken between the 12th and 13th ribs. The left side of al l carcasses was dissected into muscle, subcutaneous fat, intermuscular fat, bone, and remainder (major blood vessels, ligaments, tendons, and thick connective tissue sheets associated with muscles). The carcasses lean meat percentage (LMP), total fat percentage (FP), and bone percentage (BP) were calculated. A SEM model was fited by OLS, WLS and SUR estimators. Models fitting quality was evaluated by the coefficient of determination, the root mean square error, and Log-likelihood statistic. This study shows that SUR estimates are consistently better than the OLS and WLS estimates for modeling the carcass composition of lambs...

A Spatial and Temporal Prediction Model of Corn Grain Yield as a Function of Soil Attributes

Rodrigues, Marcos S.; Cora, Jose E.; Castrignano, Annamaria; Mueller, Tom G.; Rienzi, Eduardo
Fonte: Amer Soc Agronomy Publicador: Amer Soc Agronomy
Tipo: Artigo de Revista Científica Formato: 1878-1887
Português
Relevância na Pesquisa
36.3%
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES); Effective site-specific management requires an understanding of the soil and environmental factors influencing crop yield variability. Moreover, it is necessary to assess the techniques used to define these relationships. The objective of this study was to assess whether statistical models that accounted for heteroscedastic and spatial-temporal autocorrelation were superior to ordinary least squares (OLS) models when evaluating the relationship between corn (Zea mays L.) yield and soil attributes in Brazil. The study site (10 by 250 m) was located in Sao Paulo State, Brazil. Corn yield (planted with 0.9-m spacing) was measured in 100 4.5- by 10-m cells along four parallel transects (25 observations per transect) during six growing seasons between 2001 and 2010. Soil chemical and physical attributes were measured. Ordinary least squares, generalized least squares assuming heteroscedasticity (GLS(he)), spatial-temporal least squares assuming homoscedasticity (GLS(sp)), and spatial-temporal assuming heteroscedasticity (GLS(he-sp)) analyses were used to estimate corn yield. Soil acidity (pH) was the factor that most influenced corn yield with time in this study. The OLS model suggested that there would be a 0.59 Mg ha(-1) yield increase for each unit increase in pH...

Money demand in the euro area, the US and the UK : assessing the role of nonlinearity

Jawadi, Fredj; Sousa, Ricardo M.
Fonte: Universidade do Minho Publicador: Universidade do Minho
Tipo: Trabalho em Andamento
Publicado em //2012 Português
Relevância na Pesquisa
26.18%
This paper estimates money demand equations for the euro area, the US and the UK using three different econometric methodologies: (i) a linear model based on a dynamic ordinary least squares (DOLS); (ii) a nonlinear technique based on a quantile regression framework; and (iii) a nonlinear model relying on a smooth-transition regression. The linear model shows that the elasticity of money demand with respect to income is positive and large in magnitude, while the elasticity of money demand with respect to the interest rate is negative and generally small. The quantile regression technique highlights that: (i) the income and the interest rate semi-elasticities are significantly different from the OLS estimates at the tails of the distribution of real money holdings; and (ii) the sensitivity of money demand with respect to inflation tends to be larger when real money holdings are extremely low. Finally, the smooth transition model provides two interesting findings. On the one hand, they capture reasonably well the dynamics of the money demand function. On the other hand, they show that the elasticity of money demand with respect to inflation rate, interest rate and GDP varies not only in accordance with the regime considered...

Financial market and the macroeconomic variables

Gomes, Carla Cindy Mendes
Fonte: Instituto Superior de Economia e Gestão Publicador: Instituto Superior de Economia e Gestão
Tipo: Dissertação de Mestrado
Publicado em //2013 Português
Relevância na Pesquisa
26.25%
Mestrado em Finanças; This study aims to examine the effect of the macroeconomic variables on the stock market price index from Germany and Portugal, using the OLS regression model and quarterly data from 2000(Q1) to 2011(Q4). The group of the macroeconomic variables used in this study is composed by GDP, consumer price index, long term domestic interest rate, exchange rate, and by the ratio of government deficit, tax revenue, net lending or borrowing of an economy and gross fixed capital formation, to GDP. In addition to the macroeconomic variables presented, we also consider the Dow Jones Industrial Average price index and the US long term interest rate. Considering all the explanatory variables on the regression model, we found that both stock markets analyzed are positively influenced by Dow Jones return and US long term interest rate change, and negatively affected by the depreciation of the exchange rate. Germany stock return is positively affected by the domestic long run interest rate change. In regards to the Portugal stock return, it is positively influenced by the GDP growth rate and negatively affected by the growth rate of the consumer price index. Concerning the policy implication, to promote a robust stock market, the authorities are expected to manage the domestic interest rate...

The role of macroeconomics in the portuguese stock market

Gonçalves, Paulo José Ribeiro
Fonte: Instituto Universitário de Lisboa Publicador: Instituto Universitário de Lisboa
Tipo: Dissertação de Mestrado
Publicado em //2012 Português
Relevância na Pesquisa
46.33%
Mestrado em Finanças; Este estudo investiga a relação entre variáveis macroeconómicas e o retorno das ações (PSI 20 e suas empresas), usando dados mensais que variam de Janeiro de 1999 a Novembro de 2011. As variáveis macroeconómicas utilizadas neste estudo são o índice de preços no consumidor (como uma proxy para a inflação), índice de produção industrial, taxa de câmbio (EUR/USD), taxas de juro (taxa de juro a 10 anos e EURIBOR de três meses) e agregado monetário (M2). O modelo de estimação dos mínimos quadrados ordinários (OLS) foi utilizado para estabelecer a relação entre variáveis macroeconómicas e retornos do mercado de ações. Os resultados empíricos revelam que existem alguns casos em que se verifica uma relação estatisticamente significativa entre retornos das acções e nossas variáveis macroeconómicas. Conclui-se ainda que as variáveis macroeconómicas afectam os retornos do PSI 20 e as suas empresas da mesma forma. Os resultados por nós obtidos podem ainda fornecer algumas indicações a gerentes de empresas, investidores e corretores.; This study investigates the relation between macroeconomic variables and stock market returns (PSI 20 index and its companies) using monthly data that ranging from January 1999 to November 2011. Macroeconomic variables used in this study are consumer price index (as a proxy for inflation)...

Prediction of hybrid means from a partial circulant diallel table using the ordinary least square and the mixed model methods

Reis,Américo José dos Santos; Chaves,Lázaro José; Duarte,João Batista; Brasil,Edward Madureira
Fonte: Sociedade Brasileira de Genética Publicador: Sociedade Brasileira de Genética
Tipo: Artigo de Revista Científica Formato: text/html
Publicado em 01/01/2005 Português
Relevância na Pesquisa
26.2%
By definition, the genetic effects obtained from a circulant diallel table are random. However, because of the methods of analysis, those effects have been considered as fixed. Two different statistical approaches were applied. One assumed the model to be fixed and obtained solutions through the ordinary least square (OLS) method. The other assumed a mixed model and estimated the fixed effects (BLUE) by generalized least squares (GLS) and the best linear unbiased predictor (BLUP) of the random effects. The goal of this study was to evaluate the consequences when considering these effects as fixed or random, using the coefficient of correlation between the responses of observed and non-observed hybrids. Crossings were made between S1 inbred lines from two maize populations developed at Universidade Federal de Goiás, the UFG-Samambaia "Dent" and UFG-Samambaia "Flint". A circulant inter-group design was applied, and there were five (s = 5) crossings for each parent. The predictions were made using a reduced model. Diallels with different sizes of s (from 2 to 5) were simulated, and the coefficients of correlation were obtained using two different approaches for each size of s. In the first approach, the observed hybrids were included in both the estimation of the genetic parameters and the coefficient of correlation...

Human monoclonal antibodies reactive to oligodendrocytes promote remyelination in a model of multiple sclerosis

Warrington, Arthur E.; Asakura, Kunihiko; Bieber, Allan J.; Ciric, Bogoljub; Van Keulen, Virginia; Kaveri, Srini V.; Kyle, Robert A.; Pease, Larry R.; Rodriguez, Moses
Fonte: National Academy of Sciences Publicador: National Academy of Sciences
Tipo: Artigo de Revista Científica
Publicado em 06/06/2000 Português
Relevância na Pesquisa
26.25%
Promoting remyelination, a major goal of an effective treatment for demyelinating diseases, has the potential to protect vulnerable axons, increase conduction velocity, and improve neurologic deficits. Strategies to promote remyelination have focused on transplanting oligodendrocytes (OLs) or recruiting endogenous myelinating cells with trophic factors. Ig-based therapies, routinely used to treat a variety of neurological and autoimmune diseases, underlie our approach to enhance remyelination. We isolated two human mAbs directed against OL surface antigens that promoted significant remyelination in a virus-mediated model of multiple sclerosis. Four additional OL-binding human mAbs did not promote remyelination. Both human mAbs were as effective as human i.v. Ig, a treatment shown to have efficacy in multiple sclerosis, and bound to the surface of human OLs suggesting a direct effect of the mAbs on the cells responsible for myelination. Alternatively, targeting human mAbs to areas of central nervous system (CNS) pathology may facilitate the opsonization of myelin debris, allowing repair to proceed. Human mAbs were isolated from the sera of individuals with a form of monoclonal gammopathy. These individuals carry a high level of monoclonal protein in their blood without detriment...

NMDA Receptor Blockade with Memantine Attenuates White Matter Injury in a Rat Model of Periventricular Leukomalacia

Manning, Simon M.; Talos, Delia M.; Zhou, Chengwen; Selip, Debra B.; Park, Hyun-Kyung; Park, Chang-Joo; Volpe, Joseph J.; Jensen, Frances E.
Fonte: PubMed Publicador: PubMed
Tipo: Artigo de Revista Científica
Publicado em 25/06/2008 Português
Relevância na Pesquisa
26.36%
Hypoxia–ischemia (H/I) in the premature infant leads to white matter injury termed periventricular leukomalacia (PVL), the leading cause of subsequent neurological deficits. Glutamatergic excitotoxicity in white matter oligodendrocytes (OLs) mediated by cell surface glutamate receptors (GluRs) of the AMPA subtype has been demonstrated as one factor in this injury. Recently, it has been shown that rodent OLs also express functional NMDA GluRs (NMDARs), and overactivation of these receptors can mediate excitotoxic OL injury. Here we show that preterm human developing OLs express NMDARs during the PVL period of susceptibility, presenting a potential therapeutic target. The expression pattern mirrors that seen in the immature rat. Furthermore, the uncompetitive NMDAR antagonist memantine attenuates NMDA-evoked currents in developing OLs in situ in cerebral white matter of immature rats. Using an H/I rat model of white matter injury, we show in vivo that post-H/I treatment with memantine attenuates acute loss of the developing OL cell surface marker O1 and the mature OL marker MBP(myelin basic protein), and also prevents the long-term reduction in cerebral mantle thickness seen at postnatal day 21 in this model. These protective doses of memantine do not affect normal myelination or cortical growth. Together...

AMP-Activated Protein Kinase Signaling Protects Oligodendrocytes that Restore Central Nervous System Functions in an Experimental Autoimmune Encephalomyelitis Model

Paintlia, Ajaib S.; Paintlia, Manjeet K.; Mohan, Sarumathi; Singh, Avtar K.; Singh, Inderjit
Fonte: American Society for Investigative Pathology Publicador: American Society for Investigative Pathology
Tipo: Artigo de Revista Científica
Publicado em /08/2013 Português
Relevância na Pesquisa
26.42%
AMP-activated protein kinase (AMPK) signaling is reported to protect neurons under pathologic conditions; however, its effect on oligodendrocytes (OLs) remains to be elucidated. We investigated whether AMPK signaling protects OLs to restore central nervous system (CNS) functions in an experimental autoimmune encephalomyelitis (EAE), a murine model of multiple sclerosis. Increased inflammation and demyelination in the CNS and peripheral immune responses were consistent with the observed clinical impairments in EAE animals, which were attenuated by treatment with metformin compared with vehicle. In addition, expressions of neurotrophic factors and of signatory genes of OL lineages were increased in the CNS of metformin-treated EAE animals. Likewise, metformin attenuated inflammatory response and enhanced expressions of neurotrophic factors, thereby protecting OLs via AMPK activation in mixed glial cultures stimulated with lipopolysaccharide/interferon γ in vitro, as evidenced by analysis of the expression of signatory genes of O1+/MBP+ OLs and their cellular populations. Metformin also attenuated oxidative stress and malondialdehyde-containing protein levels, with corresponding induction of antioxidative defenses in OLs exposed to cytokines via AMPK activation. These effects of metformin were evident in the CNS of EAE animals. These data provide evidence that AMPK signaling is crucial to protect OLs and...

Evaluation of NPP-VIIRS Nighttime Light Data for Mapping Global Fossil Fuel Combustion CO2 Emissions: A Comparison with DMSP-OLS Nighttime Light Data

Ou, Jinpei; Liu, Xiaoping; Li, Xia; Li, Meifang; Li, Wenkai
Fonte: Public Library of Science Publicador: Public Library of Science
Tipo: Artigo de Revista Científica
Publicado em 21/09/2015 Português
Relevância na Pesquisa
26.35%
Recently, the stable light products and radiance calibrated products from Defense Meteorological Satellite Program’s (DMSP) Operational Linescan System (OLS) have been useful for mapping global fossil fuel carbon dioxide (CO2) emissions at fine spatial resolution. However, few studies on this subject were conducted with the new-generation nighttime light data from the Visible Infrared Imaging Radiometer Suite (VIIRS) sensor on the Suomi National Polar-orbiting Partnership (NPP) Satellite, which has a higher spatial resolution and a wider radiometric detection range than the traditional DMSP-OLS nighttime light data. Therefore, this study performed the first evaluation of the potential of NPP-VIIRS data in estimating the spatial distributions of global CO2 emissions (excluding power plant emissions). Through a disaggregating model, three global emission maps were then derived from population counts and three different types of nighttime lights data (NPP-VIIRS, the stable light data and radiance calibrated data of DMSP-OLS) for a comparative analysis. The results compared with the reference data of land cover in Beijing, Shanghai and Guangzhou show that the emission areas of map from NPP-VIIRS data have higher spatial consistency of the artificial surfaces and exhibit a more reasonable distribution of CO2 emission than those of other two maps from DMSP-OLS data. Besides...

The estimation of the rate of return to education in China: an empirical analysis using instrument variable estimation with months of birth and its Issues

Cowling, Michael Leith
Fonte: Universidade Nacional da Austrália Publicador: Universidade Nacional da Austrália
Tipo: Thesis (Honours)
Português
Relevância na Pesquisa
26.38%
Any attempt to estimate the rate of return to education using ordinary least squares (OLS) models suffers from omitted variable bias due to unobservable factors that are correlated with both the education variable and the return dependent variable. Instrument Variables, such as the birth months of students, provide an alternative estimation method that can create less biased estimates. The validity of the birth months as instrument variables depends on being uncorrelated with individual personal attributes while having an effect on the education outcome of the individual. However, the exogenous criterion is violated if unobservable factors influences the month of birth and education outcome creating the omitted variable bias problem. We investigate if the birth month is a good instrument for use in estimating the rate of return to education using empirical evidence from the 2000 Chinese Population Census and the 2009 Chinese Urban Household Income and Expenditure Survey. We split the sample into two groups, individuals with rural education and individuals with urban education due to an urban/rural education gap that the literature captures. A Two Stage Least Squares Model (TSLS) is run to estimate the rate of return to education and to determine if the instrument birth month variables are strong instruments. We also run an OLS model to compare the OLS rate of return to education with the TSLS estimates. We use the parent’s education level as a proxy for socioeconomic status and investigate if there is a violation of the exclusion restriction for the birth month instruments. We find students born after August typically achieving a higher education level on average than students born in the August and months before August. In addition...

Ols-based asymptotic inference in linear regression models with trending regressors and ar(p)-disturbances

Krämer, Walter; Mármol, Francesc
Fonte: Universidade Carlos III de Madrid Publicador: Universidade Carlos III de Madrid
Tipo: Trabalho em Andamento Formato: application/pdf
Publicado em /03/1999 Português
Relevância na Pesquisa
36.26%
We show that OLS and GLS are asymptotically equivalent in the linear regression model with AR(p)-9isturbances and a wide range of trending regressors, and that OLS-based statistical inference is still meaningful after proper adjustment of the teststatistics.

Credit default swap (CDS) prediction model & trading strategy

Dutra, Tiago Mota
Fonte: Universidade Nova de Lisboa Publicador: Universidade Nova de Lisboa
Tipo: Dissertação de Mestrado
Publicado em /01/2015 Português
Relevância na Pesquisa
35.97%
This project focuses on the study of different explanatory models for the behavior of CDS security, such as Fixed-Effect Model, GLS Random-Effect Model, Pooled OLS and Quantile Regression Model. After determining the best fitness model, trading strategies with long and short positions in CDS have been developed. Due to some specifications of CDS, I conclude that the quantile regression is the most efficient model to estimate the data. The P&L and Sharpe Ratio of the strategy are analyzed using a backtesting analogy, where I conclude that, mainly for non-financial companies, the model allows traders to take advantage of and profit from arbitrages.; UNL - NSBE

Spatial patterns of species richness in New World coral snakes and the metabolic theory of ecology

Terribile, Levi Carina; Diniz-Filho, José Alexandre Felizola
Fonte: brasil; Programa de Pós-Graduação em Ecologia e Evolução Publicador: brasil; Programa de Pós-Graduação em Ecologia e Evolução
Tipo: Artigo de Revista Científica
Português
Relevância na Pesquisa
26.28%
v. 35, p. 163-173, 2008; The metabolic theory of ecology (MTE) has attracted great interest because it proposes an explanation for species diversity gradients based on temperature-metabolism relationships of organisms. Here we analyse the spatial richness pattern of 73 coral snake species from the New World in the context of MTE. We first analysed the association between lntransformed richness and environmental variables, including the inverse transformation of annual temperature (1/kT ). We used eigenvector-based spatial filtering to remove the residual spatial autocorrelation in the data and geographically weighted regression to account for non-stationarity in data. In a model I regression (OLS), the observed slope between ln-richness and 1/kT was 0.626 (r 2 ¼ 0.413), but a model II regression generated a much steeper slope (0.975). When we added additional environmental correlates and the spatial filters in the OLS model, the R2 increased to 0.863 and the partial regression coefficient of 1/kT was 0.676. The GWR detected highly significant non-stationarity, in data, and the median of local slopes of ln-richness against 1/kT was 0.38. Our results expose several problems regarding the assumptions needed to test MTE: although the slope of OLS fell within that predicted by the theory and the dataset complied with the assumption of temperature-independence of average body size...

The Impact of Outliers on Net-Benefit Regression Model in Cost-Effectiveness Analysis

Wen, Yu-Wen; Tsai, Yi-Wen; Wu, David Bin-Chia; Chen, Pei-Fen
Fonte: Public Library of Science Publicador: Public Library of Science
Tipo: Artigo de Revista Científica
Publicado em 19/06/2013 Português
Relevância na Pesquisa
26.31%
Ordinary least square (OLS) in regression has been widely used to analyze patient-level data in cost-effectiveness analysis (CEA). However, the estimates, inference and decision making in the economic evaluation based on OLS estimation may be biased by the presence of outliers. Instead, robust estimation can remain unaffected and provide result which is resistant to outliers. The objective of this study is to explore the impact of outliers on net-benefit regression (NBR) in CEA using OLS and to propose a potential solution by using robust estimations, i.e. Huber M-estimation, Hampel M-estimation, Tukey's bisquare M-estimation, MM-estimation and least trimming square estimation. Simulations under different outlier-generating scenarios and an empirical example were used to obtain the regression estimates of NBR by OLS and five robust estimations. Empirical size and empirical power of both OLS and robust estimations were then compared in the context of hypothesis testing.

Differentially Private Ordinary Least Squares: $t$-Values, Confidence Intervals and Rejecting Null-Hypotheses

Sheffet, Or
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Português
Relevância na Pesquisa
26.33%
Linear regression is one of the most prevalent techniques in data analysis. Given a large collection of samples composed of features $\vec x$ and a label $y$, linear regression is used to find the best prediction of the label as a linear combination of the features. However, it is also common to use linear regression for its \emph{explanatory} capabilities rather than label prediction. Ordinary Least Squares (OLS) is often used in statistics to establish a correlation between an attribute (e.g. gender) and a label (e.g. income) in the presence of other features. Under the assumption of a certain random generative model for the data, OLS derives \emph{$t$-values} --- representing the likelihood of each real value to be the true correlation in the underlying distribution. Using $t$-values, OLS can release a \emph{confidence interval} that is likely to contain the true correlation. When this interval does not intersect the origin, we can \emph{reject the null hypothesis} as it is likely that $x_j$ indeed has a non-zero correlation with $y$. Our work aims at achieving similar guarantees on data under differentially private estimators. We use the Gaussian Johnson-Lindenstrauss transform, which has been shown to satisfy differential privacy if the given data has large singular values. We analyze the result of projecting the data using JLT under the OLS model and derive approximated $t$-values...

Least squares after model selection in high-dimensional sparse models

Belloni, Alexandre; Chernozhukov, Victor
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Português
Relevância na Pesquisa
26.48%
In this article we study post-model selection estimators that apply ordinary least squares (OLS) to the model selected by first-step penalized estimators, typically Lasso. It is well known that Lasso can estimate the nonparametric regression function at nearly the oracle rate, and is thus hard to improve upon. We show that the OLS post-Lasso estimator performs at least as well as Lasso in terms of the rate of convergence, and has the advantage of a smaller bias. Remarkably, this performance occurs even if the Lasso-based model selection "fails" in the sense of missing some components of the "true" regression model. By the "true" model, we mean the best s-dimensional approximation to the nonparametric regression function chosen by the oracle. Furthermore, OLS post-Lasso estimator can perform strictly better than Lasso, in the sense of a strictly faster rate of convergence, if the Lasso-based model selection correctly includes all components of the "true" model as a subset and also achieves sufficient sparsity. In the extreme case, when Lasso perfectly selects the "true" model, the OLS post-Lasso estimator becomes the oracle estimator. An important ingredient in our analysis is a new sparsity bound on the dimension of the model selected by Lasso...

A comparison of simultaneous autoregressive and generalized least squares models for dealing with spatial autocorrelation

Beguería, Santiago; Pueyo, Yolanda
Fonte: John Wiley & Sons Publicador: John Wiley & Sons
Tipo: Artículo Formato: 786717 bytes; application/pdf
Português
Relevância na Pesquisa
26.21%
The definitive version is available at: http://www3.interscience.wiley.com/journal/118545893/home; Aim: In their recent paper, Kissling & Carl (2008) recommended the spatial error simultaneous autorregresive model (SARerr) over ordinary least squares (OLS) for modelling species distribution. We compared these models with the generalized least squares model (GLS) and a variant of SAR (SARvario). GLS and SARvario are superior to standard implementations of SAR because the spatial covariance structure is described by a semivariogram model. Innovation: We used the complete datasets employed by Kissling & Carl (2008), with strong spatial autocorrelation, and two datasets in which the spatial structure was degraded by sample reduction and grid coarsening. GLS performed consistently better than OLS, SARerr and SARvario in all datasets, especially in terms of goodness of fit. SARvario was marginally better than SARerr in the degraded datasets. Main conclusions: GLS was more reliable than SAR-based models, so its use is recommended when dealing with spatially autocorrelated data.; CGL2005-01625/BOS (CICYT); Peer reviewed

Bayesian Hierarchical Models for Model Choice

Li, Yingbo
Fonte: Universidade Duke Publicador: Universidade Duke
Tipo: Dissertação
Publicado em //2013 Português
Relevância na Pesquisa
35.98%

With the development of modern data collection approaches, researchers may collect hundreds to millions of variables, yet may not need to utilize all explanatory variables available in predictive models. Hence, choosing models that consist of a subset of variables often becomes a crucial step. In linear regression, variable selection not only reduces model complexity, but also prevents over-fitting. From a Bayesian perspective, prior specification of model parameters plays an important role in model selection as well as parameter estimation, and often prevents over-fitting through shrinkage and model averaging.

We develop two novel hierarchical priors for selection and model averaging, for Generalized Linear Models (GLMs) and normal linear regression, respectively. They can be considered as "spike-and-slab" prior distributions or more appropriately "spike- and-bell" distributions. Under these priors we achieve dimension reduction, since their point masses at zero allow predictors to be excluded with positive posterior probability. In addition, these hierarchical priors have heavy tails to provide robust- ness when MLE's are far from zero.

Zellner's g-prior is widely used in linear models. It preserves correlation structure among predictors in its prior covariance...

Application of geographic weighted regression to establish flood-damage functions reflecting spatial variation

Chang,Ling-Fang; Lin,Chun-Hung; Su,Ming-Daw
Fonte: Water SA Publicador: Water SA
Tipo: Artigo de Revista Científica Formato: text/html
Publicado em 01/02/2008 Português
Relevância na Pesquisa
36.32%
Flood damage functions are necessary to ensure comprehensive flood-risk management. This study attempts to establish a residential flood-damage function through interviewing the residents living in the region where flood disasters occur frequently. Keelung River basin, near Taipei Metropolitan in Taiwan was selected as study area. Flood damages are related to the flood depths, which are the most commonly considered factor in previously published work. Ordinary least squares (OLS) regression was used to construct the flood-damage function at the beginning. Analytical results indicate that flood depth is the significant variable, but the spatial pattern of the residuals shows that residuals exhibit spatial autocorrelation. The Geographically Weighted Regression (GWR) Model was then applied to modify the traditional regression model, which cannot capture spatial variations, and to reduce the problem of spatial autocorrelation. The R-square value was found to increase from 0.15 to 0.24, and the spatial autocorrelation in the residuals was no longer evident. A modified OLS model with a dummy variable to capture the spatial autocorrelation pattern was also proposed for future applications. In conclusion, the residential flood damage is determined by flood depth and zone...