Niloofar Moosavi
A5034446903- Statistical Methods and Inference
- Advanced Causal Inference Techniques
- Statistical Methods and Bayesian Inference
- Soil and Unsaturated Flow
- Hydrology and Watershed Management Studies
- Bayesian Modeling and Causal Inference
- Fault Detection and Control Systems
- Environmental and Agricultural Sciences
- Neural Networks and Applications
- Soil Moisture and Remote Sensing
- Anomaly Detection Techniques and Applications
- Irrigation Practices and Water Management
- Climate change and permafrost
Umeå University
2021-2023
Forschungszentrum Jülich
2018
Abstract. In this paper, we present and analyze a novel global database of soil infiltration measurements, the Soil Water Infiltration Global (SWIG) database. total, 5023 curves were collected across all continents in SWIG These data either provided quality checked by scientists who performed experiments or they digitized from published articles. Data 54 different countries included with major contributions Iran, China, USA. addition to its extensive geographical coverage, cover research...
Abstract. In this paper, we present and analyze a global database of soil infiltration measurements, the Soil Water Infiltration Global (SWIG) database, for first time. total, 5023 curves were collected across all continents in SWIG database. These data either provided quality checked by scientists who performed experiments or they digitized from published articles. Data 54 different countries included with major contributions Iran, China, USA. addition to its spatial coverage, cover time...
Convolutional neural networks (CNN) have been successful in machine learning applications. Their success relies on their ability to consider space invariant local features. We the use of CNN fit nuisance models semiparametric estimation average causal effect a treatment. In this setting, are functions pretreatment covariates that need be controlled for. an application where we want estimate early retirement health outcome, propose control for time-structured covariates. Thus, is used when...
Important advances have recently been achieved in developing procedures yielding uniformly valid inference for a low dimensional causal parameter when high-dimensional nuisance models must be estimated. In this paper, we review the literature on and discuss costs benefits of using procedures. Naive estimation strategies based regularization, machine learning, or preliminary model selection stage finite sample distributions which are badly approximated by their asymptotic distributions. To...
Various methods have recently been proposed to estimate causal effects with confidence intervals that are uniformly valid over a set of data generating processes when high-dimensional nuisance models estimated by post-model-selection or machine learning estimators. These typically require all the confounders observed ensure identification effects. We contribute showing how semiparametric inference can be obtained in presence unobserved and models. propose uncertainty which allow for...
Important advances have recently been achieved in developing procedures yielding uniformly valid inference for a low dimensional causal parameter when high-dimensional nuisance models must be estimated. In this paper, we review the literature on and discuss costs benefits of using procedures. Naive estimation strategies based regularisation, machine learning, or preliminary model selection stage finite sample distributions which are badly approximated by their asymptotic distributions. To...
Convolutional neural networks (CNN) have been successful in machine learning applications. Their success relies on their ability to consider space invariant local features. We the use of CNN fit nuisance models semiparametric estimation average causal effect a treatment. In this setting, are functions pre-treatment covariates that need be controlled for. an application where we want estimate early retirement health outcome, propose control for time-structured covariates. Thus, is used when...