检索结果-内蒙古大学图书馆

generalized expectile regression with flexible response function

BIOMETRICAL JOURNAL 2021年第5期63卷 1028-1051页

作者： Spiegel, Elmar Kneib, Thomas von Gablenz, Petra Otto-Sobotka, Fabian Helmholtz Zentrum Munchen GmbH German Res Ctr Environm Hlth Inst Computat Biol Ingolstadter Landstr 1 D-85764 Neuherberg Germany Univ Goettingen Chair Stat Gottingen Germany Jade Univ Appl Sci Inst Hearing Technol & Audiol Oldenburg Germany Carl von Ossietzky Univ Oldenburg Div Epidemiol & Biometry Oldenburg Germany

Expectile regression, in contrast to classical linear regression, allows for heteroscedasticity and omits a parametric specification of the underlying distribution. This model class can be seen as a quantile-like generalization of least squares regression. Similarly as in quantile regression, the whole distribution can be modeled with expectiles, while still offering the same flexibility in the use of semiparametric predictors as modern mean regression. However, even with no parametric assumption for the distribution of the response in expectile regression, the model is still constructed with a linear relationship between the fitted value and the predictor. If the true underlying relationship is nonlinear then severe biases can be observed in the parameter estimates as well as in quantities derived from them such as model predictions. We observed this problem during the analysis of the distribution of a self-reported hearing score with limited range. Classical expectile regression should in theory adhere to these constraints, however, we observed predictions that exceeded the maximum score. We propose to include a response function between the fitted value and the predictor similarly as in generalized linear models. However, including a fixed response function would imply an assumption on the shape of the underlying distribution function. Such assumptions would be counterintuitive in expectile regression. Therefore, we propose to estimate the response function jointly with the covariate effects. We design the response function as a monotonically increasing P-spline, which may also contain constraints on the target set. This results in valid estimates for a self-reported listening effort score through nonlinear estimates of the response function. We observed strong associations with the speech reception threshold.

关键词： distributional regression generalized additive models monotonicity constraints P‐ spline single index models

来源：评论

学校读者我要写书评

暂无评论

Spatial patterns in chlorophyll a concentration during the winter-spring - spring periods in the Barents Sea

引用

PROGRESS IN OCEANOGRAPHY 2024年 228卷

作者： Dvoretsky, Vladimir G. Vodopianova, Veronika V. Bulavina, Aleksandra S. Pastukhov, Ivan A. Russian Acad Sci MMBI RAS Murmansk Marine Biol Inst Moscow Russia

Climatic fluctuations have been documented to strongly affect Arctic marine ecosystems. Plankton assemblages serve as the most sensitive indicators of such environmental forcing. We conducted a study to investigate the spatial variability of chlorophyll a (Chl-a) concentration during two pre-bloom periods (March-April 2021 and February-March 2022) in relation to the distribution of different water masses and associated properties. The upper 50 m layer of the water column was homogeneous and stable, characterized by high nutrient concentrations. Our mapping of the Barents Sea based on Chl-a concentrations revealed low estimates during the winter period. In contrast, two distinct Chl-a peaks were observed in the spring. The first region with high Chl-a concentrations was identified in Murmansk Coastal Water and Atlantic Water (0.7-1.4 mg m-- 3 ), reflecting the positive impact of the frontal zone between these interacting water masses. The second region with elevated Chla concentrations (0.9-1.1 mg m- 3 ) was located in Kolguev-Pechora Water near the southeastern ice edge. Cold water regions (Barents Sea Water, Arctic Water, Novaya Zemlya Coastal Water) exhibited low spring Chl-a concentrations (0.03-0.3 mg m-3).-3 ). generalized additive models identified hydrological variables (temperature and salinity), dissolved oxygen content, and nutrient concentrations (nitrite, nitrate, phosphate) as significant predictors explaining a substantial portion of the Chl-a variability.

关键词： Phytoplankton Chlorophyll a Nutrients Arctic shelf Barents Sea generalized additive models

来源：评论

学校读者我要写书评

暂无评论

Associations Between Anthropogenic Factors, Meteorological Factors, and Cause-Specific Emergency Department Admissions

引用

GEOHEALTH 2024年第9期8卷 e2024GH001061页

作者： Tewari, Pranav Xu, Baihui Pei, Ma Tan, Kelvin Bryan Abisheganaden, John Yim, Steve Hung-Lam Dickens, Borame Lee Lim, Jue Tao Nanyang Technol Univ Lee Kong Chian Sch Med Singapore Singapore Natl Univ Singapore Saw Swee Hock Sch Publ Hlth Singapore Singapore Minist Hlth Singapore Singapore Tan Tock Seng Hosp Singapore Singapore Nanyang Technol Univ Asian Sch Environm Singapore Singapore

Unpredictable emergency department (ED) admissions challenge healthcare systems, causing resource allocation inefficiencies. This study analyses associations between air pollutants, meteorological factors, and 2,655,861 cause-specific ED admissions from 2014 to 2018 across 12 categories. generalized additive models were used to assess non-linear associations for each exposure, yielding Incidence Rate Ratios (IRR), while the population attributable fraction (PAF) calculated each exposure's contribution to cause-specific ED admissions. IRRs revealed increased risks of ED admissions for respiratory infections (IRR: 1.06, 95% CI: 1.01-1.11) and infectious and parasitic diseases (IRR: 1.09, 95% CI: 1.03-1.15) during increased rainfall (13.21-16.97 mm). Wind speeds >12.73 km/hr corresponded to increased risks of ED admissions for respiratory infections (IRR: 1.12, 95% CI: 1.03-1.21) and oral diseases (IRR: 1.58, 95% CI: 1.31-1.91). Higher concentrations of air pollutants were associated with elevated risks of cardiovascular disease (IRR: 1.16, 95% CI: 1.05-1.27 for PM10) and respiratory infection-related ED admissions (IRR: 2.78, 95% CI: 1.69-4.56 for CO). Wind speeds >12.5 km/hr were predicted to contribute toward 10% of respiratory infection ED admissions, while mean temperatures >28 degrees C corresponded to increases in the PAF up to 5% for genitourinary disorders and digestive diseases. PM10 concentrations >60 mu g/m(3) were highly attributable toward cardiovascular disease (PAF: 10%), digestive disease (PAF: 15%) and musculoskeletal disease (PAF: 10%) ED admissions. CO concentrations >0.6 ppm were highly attributable to respiratory infections (PAF: 20%) and diabetes mellitus (PAF: 20%) ED admissions. This study underscores protective effects of meteorological variables and deleterious impacts of air pollutant exposures across the ED admission categories considered. Plain Language Summary The unpredictability and variability of emergency department (ED) admissions lo

关键词： non-linear associations generalized additive models population attributable fraction Singapore emergency department admissions

来源：评论

学校读者我要写书评

暂无评论

Detecting ecological signatures of long-term human activity across an elevational gradient in the Sumava Mountains, Central Europe

引用

QUATERNARY SCIENCE REVIEWS 2024年 344卷

作者： Kraklow, Vachel A. Dreslerova, Dagmar Diaconu, Andrei-Cosmin Moravcova, Alice Kadlec, Martin Nyvlt, Daniel Tinner, Willy Heurich, Marco Finsinger, Walter Feurdean, Angelica Kunes, Petr Florescu, Gabriela Charles Univ Prague Dept Bot Prague Czech Republic Univ Utah Dept Geog Salt Lake City UT USA Los Alamos Natl Lab Earth & Environm Sci Div Los Alamos NM USA Czech Acad Sci Inst Archaeol Prague Czech Republic Babes Bolyai Univ Dept Geol Kogalniceanu 1 Cluj Napoca 400084 Romania Masaryk Univ Dept Geog Brno Czech Republic Univ Bern Inst Plant Sci Bern Switzerland Univ Bern Oeschger Ctr Climate Change Res Bern Switzerland Bavarian Forest Natl Pk Dept Natl Pk Monitoring & Anim Management Freyunger Str 2 D-94481 Grafenau Germany Univ Freiburg Fac Environm & Nat Resources Chair Wildlife Ecol & Management Tennenbacher Str 4 D-79106 Freiburg Germany Inland Norway Univ Appl Sci Inst Forestry & Wildlife Management NO-2480 Koppang Norway Univ Montpellier ISEM CNRS IRD Montpellier France Goethe Univ Dept Phys Geog Altenhoferallee 1 D-60438 Frankfurt Germany Stefan Cel Mare Univ Integrated Ctr Res Dev & Innovat Adv Mat Nanotechnol & Distributed Syst Fabricat & Control Univ 13 Suceava 720229 Romania

Central European mountains, including the Sumava Mountains located along the Czechia/Germany border, have a long and rich anthropogenic history. Yet, documenting prehistoric human impact in Central European mountain environments remains a challenge because of the need to disentangle climate and human-caused responses in terrestrial systems. Here, we present the first reconstructed water table depths (WTDs) from two sites, Pekna and Blatenska slat, located in the Sumava Mountains. We compare these local WTD records with new and published pollen, non-pollen palynomorphs (NPPs), plant macrofossils, geochemistry and archeological records to investigate how changes in local hydrology and human activities impacted forest succession and fire activity throughout the Holocene across an elevational gradient. Using a generalized additive model, our results suggest that changes in forest succession and fire activity have been primarily caused by climate throughout the Holocene. However, humans have been utilizing mountain environments and their resources continuously since similar to 4600 cal yr BP, thus playing a secondary role in modifying forest succession to increase resources beneficial to both humans and grazers. Over the last 1000 years, we provide evidence of directly observed human-caused modifications to the landscape. These results contribute to a growing body of literature illustrating human activities and landscape modifications in Central European mountains.

关键词： Peat bog Testate amoebae Water table depth Fire Pollen Plant macrofossils Climate change Human impact Holocene generalized additive models

来源：评论

学校读者我要写书评

暂无评论

The optimal realized niches of the diatom Paralia sulcata in the Yellow Sea and its environmental implication

引用

ECOLOGICAL INDICATORS 2024年 158卷

作者： He, Mengfan Zhou, Xinyu Lv, Ting Wang, Yueqi Liu, Dongyan Wang, Yujue East China Normal Univ Inst Ecochongming State Key Lab Estuarine & Coastal Res Shanghai 200241 Peoples R China Chinese Acad Sci Yantai Inst Coastal Zone Res Key Lab Coastal Zone Environm Proc & Ecol Remediat Yantai 264003 Peoples R China

The observational and paleo-ecological records on some shelves (e.g., North Sea, Yellow Sea) have indicated that the proportion of Paralia sulcata in the diatom community has been increasing since the 1980s. Because of its heavily-silicified frustules, this species is considered as an important siliceous organism, which can impact the recycling efficiency of biogenic silicon in the ocean. However, the specific environmental factors driving the increase of P. sulcata remain unclear. In this study, based on seasonal in-situ observations and historical literature in the Yellow Sea, we statistically assessed the importance of different environmental factors in determining P. sulcata abundance. Further, we quantified its optimal realized niches, using the maximum entropy and generalized additive models to predict the fate of this species. P. sulcata indicates a preference for environments with cold temperatures and high nutrient levels. The contribution of sea temperature on cell abundance reaches up to 37.2 %, which is higher than that of the nutrient concentrations (-20.5 %) and salinity (-17.1 %). The optimal realized niches are defined as sea temperature at 6.25 +/- 3.50 degrees C, salinity at 32.8 +/- 0.52, the concentrations of dissolved inorganic nitrogen, phosphorus, and silicate at 8.38 +/- 5.42 mu M, 0.42 +/- 0.29 mu M, and 5.87 +/- 3.53 mu M, respectively, and ammonium/nitrate ratio at 1.32 +/- 0.19. Based on the multivariate statistical methods, the optimal realized niches indicate that the increased sea temperatures in winter and spring, combined with nitrogen enrichment, especially ammonium enrichment, are key environmental drivers for P. sulcata proliferation in the Yellow Sea.

关键词： Diatom MaxEnt model generalized additive models Ocean warming Nitrogen enrichment

来源：评论

学校读者我要写书评

暂无评论

Association between ambient temperature and thyroid-stimulating hormone and free thyroxine levels in Korean euthyroid adults

引用

ENVIRONMENTAL RESEARCH 2024年第Part2期262卷 119918页

作者： Sung, Jisun Kim, Jong-Hun Sungkyunkwan Univ Dept Social & Prevent Med Sch Med 2066 Seobu Ro Suwon Gyeonggi Do South Korea

Thyroid-stimulating hormone (TSH) and free thyroxine (FT4) levels are influenced by factors such as genetics, lifestyle, and seasonal variations. This study aimed to investigate the association between ambient temperature and TSH and FT4 levels in Korean adults with normal thyroid function. Data from 4659 euthyroid adults who participated in the 6th Korea National Health and Nutrition Examination Survey (KNHANES) from 2013 to 2015 were analyzed. Meteorological data linked to the KNHAES data based on the participants' residential addresses and examination dates were used to estimate individual temperature exposure. generalized additive models revealed a negative association between TSH and temperature, with TSH levels increasing by 4.5% (95% confidence interval [CI] 1.1%, 7.9%) at extreme cold (1st percentile temperature) and decreasing by- 3.7% (95% CI -6.4%,- 0.9%) at extreme heat (99th percentile temperature) compared to the 49th percentile temperature. FT4 levels showed no significant association with cold temperatures but decreased by- 1.4% (95% CI -2.6%,- 0.2%) at extreme heat (99th percentile temperature) compared to the 85th percentile temperature. TSH responded sensitively across the entire temperature range, whereas FT4 responded only to heat. Our study showed the physiological responses and adaptation of thyroid function to extreme temperatures, providing insights into health research in the context of a changing climate.

关键词： BMI Body mass index FT4 Free thyroxine FT3 Free triiodothyronine GAM generalized additive models KNHANES Korea National Health and Nutrition Examination Survey TSH Thyroid-stimulating hormone TRH Thyrotropin-releasing hormone T4 Thyroxine T3 Triiodothyronine TPOAb Thyroid peroxidase antibody

来源：评论

学校读者我要写书评

暂无评论

Mixed stands of black spruce (Picea mariana) and tamarack (Larix laricina) offer high secondary growth in eastern boreal forests of Canada

引用

FOREST ECOLOGY AND MANAGEMENT 2024年 571卷

作者： Proulx, Samuel Roy Leduc, Alain Thiffault, Nelson Chavardes, Raphael D. Univ Quebeca Montreal Ctr Etud Foret CP 8888 Succ Ctr Ville Montreal PQ H3C 3P8 Canada Ctr Canadien Fibre bois Serv Canadien Forets Ressources Nat Canada 1055 rue PEPSCP 10380Succ Sainte Foy Quebec City G1V 4C7 PQ Canada Ctr Foresterie Atlantique Serv Canadien Forets Ressources Nat Canada Fredericton NB E3B 5P7 Canada Univ Quebec Abitibi Temiscamingue Inst Rech forets 445 Blvd Univ Rouyn Noranda PQ J9X 5E4 Canada Univ New Brunswick Fac Forestry & Environm Management 28 Dineen Dr Fredericton NB E3B 5A3 Canada

The Canadian boreal forest is still lacking investigation into compatible species mixtures that can lead to higher growth rates than monospecific or pure stands. The effect of species mixtures on stand productivity can vary with stem density, species proportion, site characteristics, and climate variation. We investigated mixed stands of black spruce (Picea mariana) and tamarack (Larix laricina) in eastern Canadian boreal forests. Using modern forest survey data, we analyzed stand basal area variation in 756 mature stands exclusively composed of these two species over a study area spanning 535,000 km(2). We generated a linear relationship between mixed stand basal area and stem density, and the residual of this relation was used as response variable named 'residual basal area'. Positive and negative deviations of residual basal area were analysed with generalized additive models and quantile regressions with tamarack basal area, climate, and site characteristics as explanatory variables. The selected model included tamarack basal area, mean seasonal precipitation, degree days, mean annual temperature, stand age, latitude, longitude, and the type of surficial deposit. Tamarack exhibited a linear positive effect on predicted residual basal area. The most significant effect was observed from the interaction between tamarack basal area and stand age, predicting positive residual basal area of 16 m(2) ha(-1) when stand aged was 150 years old and tamarack basal area was between 12 and 17 m(2) ha(-1). Quantile regressions revealed that tamarack could enhance basal area while increasing black spruce stem size and density. Landscape managers may consider the establishment of mixed black spruce and tamarack stands to favor secondary growth in eastern Canadian boreal forests.

关键词： Basal area variations Mixed species stands generalized additive models Mixed stands management Niche complementarity

来源：评论

学校读者我要写书评

暂无评论

Locally interpretable tree boosting: An application to house price prediction

引用

DECISION SUPPORT SYSTEMS 2024年 178卷

作者： Hjort, Anders Scheel, Ida Sommervoll, Dag Einar Pensar, Johan Univ Oslo Dept Math Oslo Norway Eiendomsverdi AS Oslo Norway Norwegian Univ Life Sci NMBU Sch Econ & Business As Norway NTNU Business Sch Trondheim Norway Postboks 1053 Blindern N-1053 Oslo Norway

We introduce Locally Interpretable Tree Boosting (LitBoost), a tree boosting model tailored to applications where the data comes from several heterogeneous yet known groups with a limited number of observations per group. LitBoost constraints the complexity of a Gradient Boosted Trees model in a way that allows us to express the final model as a set of local generalized additive models, yielding significant interpretability benefits while still maintaining some of the predictive power of a Gradient Boosted Trees model. We use house price prediction as a motivating example and demonstrate the performance of LitBoost on a data set of N = 14382 observations from 15 different city districts in Oslo (Norway). We also test the robustness of LitBoost in an extensive simulation study on a synthetic data set.

关键词： Gradient boosted trees generalized additive models Explainable boosting machines Interpretable machine learning House Price prediction

来源：评论

学校读者我要写书评

暂无评论

Differentially Private Estimation of Heterogeneous Causal Effects 1

Differentially Private Estimation of Heterogeneous Causal Ef...

引用

Conference on Causal Learning and Reasoning

作者： Niu, Fengshi Nori, Harsha Quistorff, Brian Caruana, Rich Ngwe, Donald Kannan, Aadharsh Stanford Univ Grad Sch Business Stanford CA 94305 USA Microsoft Res Redmond WA 98052 USA US Bur Econ Anal Washington DC 20233 USA

Estimating heterogeneous treatment effects in domains such as healthcare or social science often involves sensitive data where protecting privacy is important. We introduce a general meta-algorithm for estimating conditional average treatment effects (CATE) with differential privacy (DP) guarantees. Our meta-algorithm can work with simple, single-stage CATE estimators such as S-learner and more complex multi-stage estimators such as DR and R-learner. We perform a tight privacy analysis by taking advantage of sample splitting in our meta-algorithm and the parallel composition property of differential privacy. In this paper, we implement our approach using DP-EBMs as the base learner. DP-EBMs are interpretable, high-accuracy models with privacy guarantees, which allow us to directly observe the impact of DP noise on the learned causal model. Our experiments show that multi-stage CATE estimators incur larger accuracy loss than single-stage CATE or ATE estimators and that most of the accuracy loss from differential privacy is due to an increase in variance, not biased estimates of treatment effects.

关键词： causal inference differential privacy interpretability generalized additive models

来源：评论

学校读者我要写书评

暂无评论

Assessing temporal variability of lake turbidity and trophic state of European lakes using open data repositories

引用

SCIENCE OF THE TOTAL ENVIRONMENT 2023年第Pt 3期857卷 159618页

作者： Stefanidis, Konstantinos Varlas, George Papaioannou, George Papadopoulos, Anastasios Dimitriou, Elias Inst Marine Biol Resources & Inland Waters Hellen Ctr Marine Res 46-7 Km Athens Sounio Ave Anavyssos 19013 Attica Greece Univ Patras Dept Biol Univ Campus Rio GR-26500 Patras Greece Democritus Univ Thrace Dept Forestry & Management Environm & Nat Resource Orestiada 68200 Greece

Water turbidity is one of the more important water quality parameters that is strictly linked with the productivity of the lake and is commonly used as an indicator of the trophic state. However, limited field data availability across wide geo-graphic gradients may hinder the conduction of large scale longitudinal studies. In this study, time series of lake tur-bidity and trophic state index (TSI) between 2002 and 2012 were obtained from the Copernicus Lake Water products to create a large longitudinal dataset of lake variables for 22 European lakes. The dataset was combined with estimates of nutrient concentrations and surface water temperature obtained from the Hydrological Predictions for the Environment (HYPE) and ERA5-Land data repositories, that were used as environmental predictors. Hence, the validity of the lake water quality parameters was tested by a) exploring their spatial and temporal variability and b) identifying associations with the environmental predictors. For this purpose, seasonal Mann-Kendall tests were applied to find significant inter-annual trends of turbidity and TSI for each lake, and generalized additive models (GAMs) were employed to identify the main parameters that shape their temporal dynamics. Although we did not find significant inter-annual changes, our findings highlighted the strong influence of seasonality and surface water temper-ature in defining the temporal variability patterns in most of the lakes. In addition, the importance of nutrients varied among lakes as several lakes exhibited narrow nutrient gradients reflecting relatively stable nutrient conditions during the examined period. Other lake intrinsic factors, such as local climate and biotic interactions, are important drivers of shaping turbidity and nutrient dynamics. This study highlighted the usefulness of combining lake data from large repositories in conducting large scale spatial studies as a valuable asset for future lake research and management purposes.

关键词： Lakes Eutrophication Turbidity Trophic state index generalized additive models ERA5-land

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：