Sujit K. Sahu’s research while affiliated with University of Southampton and other places

... It was found that the shapes of the diagrams were close to ellipse, and they also met the assumption of multivariate normality and linearity (Çokluk et al., 2012). There is no autocorrelation between the variables tested with the Durbin-Watson (DW) statistic (DW = 2.005), and the multivariate kurtosis critical ratio was less than 10 (Küçüksille, 2014). Accordingly, maximum likelihood was used as the estimation method. ...

Reference:

Computational thinking in primary school: effects of student and school characteristics

... Instead of representing estimates as single values, these intervals provide a range that quantifies potential errors, offering a more comprehensive view of the underlying data [26]. By identifying and eliminating the causes of uncertainty, researchers can find better statistical methods with lower levels of uncertainty, thus enhancing their decision-making processes [27]. This uncertainty evaluation is essential across all scientific disciplines, as statistical methods provide the foundation for data analysis and scientific inquiry. ...

Reference:

Exploring the interdisciplinary connections between statistics and reader psychology: Insights into reading behavior

... The WHO and UNICEF produce WHO-UNICEF Estimates of National Immunization Coverage (WUENIC), synthesizing information from national reports, standardized surveys, and other credible data streams [70]. The triangulation aims to produce a more robust and comprehensive picture of vaccination efforts, correcting for potential underreporting or reporting lags. ...

Reference:

The Impact of COVID-19 on DTP3 Vaccination Coverage in Europe (2012–2023)

... Following previous work [17][18][19]27,28], we assembled a suite of geospatial covariate information for our analysis. This included remoteness (e.g., distance to roads and travel time to the nearest health facility), socioeconomic (e.g., nightlight intensity, livestock density) and environmental (e.g., elevation, temperature and precipitation) variables-see, e.g., Supplementary Table S1. ...

Reference:

Geospatial Variation in Vaccination Coverage and Zero-Dose Prevalence at the District, Ward and Health Facility Levels Before and After a Measles Vaccination Campaign in Nigeria

... Specifically, regression analysis was conducted for the effects of social and economic risk factors [13,22,27,33,40], time-series models were utilized to deal with temporal dependency throughout the pandemic [25,5] and spatial models have been employed for understanding the diffusion of COVID-19 spread [10]. Other recent studies attempted to build efficient methods that incorporate both time and space effects in order to model COVID-19's dynamics [11,28,30,38]. ...

Reference:

Temporal dynamics for areal unit-based co-occurrence COVID-19 trajectories

... Moreover, the Bayesian estimates come from the posterior predictive distributions averaging over the model's random effects. That makes the Bayesian approach is more realistic in addition to its capability to handle missing data with flexibility [1], [3]. ...

Reference:

Ride-Sharing Determinants: Spatial and Spatio-Temporal Bayesian Analysis for Chicago Service in 2022

... Spatiotemporal forecasting (and explanation) is challenging because of the characteristics of spatiotemporal data. Particularly, spatiotemporal dependence and variability, and nonlinearity in the data violate the white noise assumption for the error term (Sahu and Böhning 2021). Giacomini and Granger (2004) showed that ignoring spatial dependence in spatiotemporal forecasting can lead to highly inaccurate forecasts. ...

Reference:

Spatiotemporal forecasting models with and without a confounded covariate

... The long-term trend of Chl-a sat over the oligotrophic waters of the ocean gyres has been studied without a general agreement between increase (McClain et al., 2004;Aiken et al., 2017;Hammond et al., 2020;Cael et al., 2023) or decrease (Vantrepotte and Mélin, 2011;Signorini et al., 2015). The opposite trend observed between studies is mainly related to the spatiotemporal coverage. ...

Reference:

Spatiotemporal variability of chlorophyll-a concentration in the South Brazil Bight using 25 years of multi-sensor orbital data (1998–2022)

... When using machine learning methods in this setting, there is no obvious quantification of sampling error in most cases (Faraway & Augustin, 2018). For example, when one uses a random forest model (Breiman, 2001a), one can calculate a metric to assess the predictive performance of the model (e.g., RMSE) and extract a predictor importance metric (e.g., permutation importance; Breiman, 2001a); but in most applications, we do not see standard errors nor p values associated with these quantities (R. Sambasivan et al., 2020) 3 . ...

Reference:

Bridging Traditional Statistics and Machine Learning Approaches in Psychology: Navigating Small Samples, Measurement Error, Non-independent Observations and Missing Data

... The experts selected questions and reviewed them in terms of their social and cultural universality, overlap with the minimal ICF generic set [13] and statistical criteria. In a second step, the robustness and reliability of the selection the expert proposed were tested using Generalized Partial Credit Model (GPCM) and Bayesian models adjusted for age, gender and income were estimated [14]. ...

Reference:

Measuring functioning and disability using household surveys: metric properties of the brief version of the WHO and World Bank model disability survey