BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Institute of Statistical Research and Training - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://isrt.ac.bd
X-WR-CALDESC:Events for Institute of Statistical Research and Training
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:UTC
BEGIN:STANDARD
TZOFFSETFROM:+0000
TZOFFSETTO:+0000
TZNAME:UTC
DTSTART:20220101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=UTC:20260305T120000
DTEND;TZID=UTC:20260305T133000
DTSTAMP:20260424T015823
CREATED:20260221T064222Z
LAST-MODIFIED:20260304T100223Z
UID:8957-1772712000-1772717400@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Thursday 05 March 2026
DESCRIPTION:Title: FertiMeter: A Data-Driven Innovation to Address the Reproductive Health Crisis of Polycystic\nOvary Syndrome \nVenue\, date and time: ISRT\, 5 March 2026\, 12:15 pm \nSpeaker: K. M. Tanvir\, Lecturer\, ISRT\, University of Dhaka \nAbstract: \nBackground:\nPolycystic ovary syndrome (PCOS) affects around 12.5% of women in Bangladesh and is a major cause of infertility and pregnancy complications. Although early detection can help manage symptoms and reduce risks\, nearly 70% of women remain undiagnosed due to limited awareness and inadequate access\nto medical care.\nObjectives:\nThis study aims to develop a data-driven machine learning model that predicts the likelihood of PCOS using non-clinical features and to integrate it into a mobile application\, FertiMeter.\nMethods:\nA total of 546 participants\, including 273 women diagnosed with PCOS and 273 without PCOS\, were enrolled in the study. The CatBoost machine learning algorithm was applied to develop a predictive model for PCOS status and the model was incorporated into the FertiMeter mobile application.\nKey Findings:\nUsing eight SHAP-selected non-clinical features\, the CatBoost model achieved an average cross-validated accuracy of 86%.\nConclusions:\nApproximately 6.7 million women in Bangladesh who remain undiagnosed with PCOS can use FertiMeter mobile application to assess their likelihood of having the condition free of cost.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-23-february-2026/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20260105T140000
DTEND;TZID=UTC:20260105T150000
DTSTAMP:20260424T015823
CREATED:20260101T042836Z
LAST-MODIFIED:20260101T042836Z
UID:8479-1767621600-1767625200@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday 5 January 2026
DESCRIPTION:Title: A Moment-Based Generalization To Post-Prediction Inference \nVenue\, date and time: ISRT\, 5 January 2026\, 2 pm \nSpeaker: Awan Afiaz\, PhD candidate at the Department of Biostatistics\, University of Washington Seattle\, WA\, USA and ISRT alumnus \nAbstract: \nAs artificial intelligence (AI) and machine learning (ML) become increasingly integrated into scientific research\, investigators frequently substitute predicted outcomes for expensive or difficult-to-measure data. However\, treating these AI/ML-generated predictions as true observations can lead to biased estimates and anti-conservative inference. While high predictive accuracy is often assumed to ensure valid downstream inference\, statistical challenges in inference with predicted data (IPD) fundamentally reduce to two sources of error: bias\, when predictions systematically distort relationships among variables\, and variance\, when uncertainty from prediction models is inadequately propagated. Wang et al. (2020) introduced post-prediction inference (PostPI)\, a pioneering method that addresses this challenge by modeling the relationship between predicted and observed outcomes in a small gold-standard dataset to calibrate inference in larger unlabeled samples. PostPI has been influential in formalizing the IPD problem and demonstrating how naive approaches fail to appropriately reflect uncertainty. However\, PostPI relies on a critical assumption: that prediction errors are uncorrelated with covariates of interest. In realistic settings where prediction algorithms exhibit systematic errors related to input features\, this assumption is often violated\, leading to biased parameter estimates and inadequate error control. We revisit PostPI in light of recent methodological advances and propose a moment-based generalization that relaxes this restrictive assumption. Our extension explicitly accounts for the covariance between prediction errors and covariates by incorporating an additional correction term estimated from the labeled dataset. This approach yields unbiased point estimates under standard conditions while incorporating a simple scaling factor that appropriately reflects the contribution of relationship model uncertainty regardless of sample size allocation. Through extensive simulations across three data-generating scenarios\, we demonstrate that our method maintains nominal Type-I error rates and achieves proper coverage probability\, even when the labeled sample is substantially smaller than the unlabeled sample settings where both naive approaches and original PostPI fail. Our work illustrates the classic bias-variance trade-off inherent to IPD’s challenges and confirms that there is no free lunch when substituting predicted outcomes for true measurements.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-5-january-2026/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20251027T140000
DTEND;TZID=UTC:20251027T150000
DTSTAMP:20260424T015823
CREATED:20251025T044211Z
LAST-MODIFIED:20251025T044211Z
UID:8255-1761573600-1761577200@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday 27 October 2025
DESCRIPTION:Title: Causal Inference with Misclassified Exposure: Correcting the IPW Estimator \nVenue\, time and date: ISRT\, 2:00 pm\, October 27\, 2025 \nSpeaker: Tarikul Islam\, Lecturer\, ISRT\, University of Dhaka \nAbstract:  \nThe inverse probability weighting (IPW) estimator is widely used for estimating the average treatment effect (ATE) in causal inference. However\, the IPW estimator is prone to bias when the exposure variable is misclassified\, even if the misclassification is non-differential. In this paper\, we propose a correction for the IPW estimator using the method of moments (MoM) to account for misclassified exposure. We derive the corrected IPW estimator\, demonstrate its unbiasedness under misclassification\, and evaluate its performance through simulation studies. Furthermore\, we discuss techniques for estimating misclassification probabilities\, including scenarios with and without validation data or exposure replication. We also derive ranges of these probabilities for conducting sensitivity analyses when no gold standard is available. Additionally\, we apply the proposed method to real-world data from the Bangladesh Multiple Indicator Cluster Survey (MICS) 2019 to estimate the causal effect of the wealth index on ICT skills among women aged 15-49\, demonstrating the robustness of our method.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-27-october-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20250818T140000
DTEND;TZID=UTC:20250818T153000
DTSTAMP:20260424T015823
CREATED:20250814T042935Z
LAST-MODIFIED:20250814T043120Z
UID:8100-1755525600-1755531000@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday August 18\, 2025
DESCRIPTION:Title: Sensor Data Analytics with Wearables and mm-Wave Radars for Digital Health \nVenue\, time and date: ISRT\, 2:00 pm\, August 18\, 2025 \nSpeaker: Shekh Md Mahmudul Islam\, Ph.D.\, Associate Professor in the Department of Electrical and Electronic Engineering at the University of Dhaka and Postdoctoral Fellow in the Department of Biomedical Engineering at Duke University\, USA. \nAbstract:  \nThis talk presents an integrated view of sensor data analytics for digital health\, combining wearable sensing and millimeter-wave (mm-Wave) radar technologies. Beyond hardware innovation\, the focus is on transforming raw physiological signals into actionable insights through advanced signal processing\, feature engineering\, and machine learning. Case studies will include digital biomarker extraction for diabetes monitoring\, continuous identity authentication\, and unobtrusive sleep apnea detection. Key topics include multi-sensor fusion of respiration and heart rate variability (HRV) data\, robust analytics pipelines resilient to noise and motion artifacts\, and the use of interpretable machine learning models for health status prediction. The presentation also covers compliance tracking using the radar sensor framework\, validation under real-world and adversarial conditions\, and the translation of analytics into clinical and remote monitoring applications. The talk concludes with opportunities for collaborative research in scaling sensor analytics for aging populations and chronic disease management.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-august-18-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20250505T140000
DTEND;TZID=UTC:20250505T153000
DTSTAMP:20260424T015823
CREATED:20250406T053746Z
LAST-MODIFIED:20250501T071751Z
UID:7545-1746453600-1746459000@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday May 5\, 2025
DESCRIPTION:Title: Understanding Causality: Rubin’s Potential Outcome Model and Philosophical Perspectives \nVenue\, time and date: ISRT\, 2:00 pm\, May 5\, 2025 \nSpeaker: Nahian Nujhat\, Bushra Chowdhury\, Md. Mutasim Billah\, Faria Rauf Ria\, Maliha Binte Alauddin\, Institute of Statistical Research and Training\, University of Dhaka. \nAbstract:  \nCausal inference focuses on estimating cause-and-effect relationships\, a key challenge in statistics\, where distinguishing association from causation is crucial. In this talk\, we will explore the foundations of causal inference through the lens of Paul W. Holland’s seminal 1986 Journal of the American Statistical Association paper “Statistics and Causal Inference.” \nRubin’s model formalizes causal effects through the potential outcomes framework\, which requires observing both counterfactuals. The two potential outcomes for a unit refer to the outcome that would be observed if the unit receives the treatment and the outcome that would be observed if the unit does not receive the treatment. The fundamental problem of causal inference is–the impossibility of observing both potential outcomes for the same unit\, which can be overcome under some untestable assumptions. \nSeveral philosophers\, such as Hume\, Mill\, and Suppes\, have contributed to understanding causation. Hume emphasized that causation is observed through temporal succession\, contiguity and constant conjunction rather than direct observation\, which led him to be skeptical about causality. Mill believed that experimental inquiry is required to identify causal relationships. Suppes advanced the discussion by introducing a probabilistic theory of causality. These philosophical views are explored in the context of Rubin’s model. \nFinally\, we will address the question of what can be a cause\, arguing that only manipulable factors can be considered causes in the context of experiments. This will lead to a discussion of the limitations of causal inference in observational studies and the importance of distinguishing between attributes and causes.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-april-07-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20250310T120000
DTEND;TZID=UTC:20250310T133000
DTSTAMP:20260424T015823
CREATED:20250306T035154Z
LAST-MODIFIED:20250306T035154Z
UID:7503-1741608000-1741613400@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday March 10\, 2025
DESCRIPTION:Title: Variational Autoencoder Model for Exploring Latent Spaces in High-Dimensional Datasets \nVenue\, time and date: ISRT\, 12:15 pm\, March 10\, 2025 \nSpeaker: Mashfiqul Huq Chowdhury\, PhD\, Associate Professor at Mawlana Bhashani Science and Technology University \nAbstract:  \nIn this talk\, I will focus on a probabilistic generative model known as the Variational Autoencoder (VAE). The VAE model uses variational Bayes to approximate the intractable posterior distribution over latent variables. I will begin by presenting the derivation of the evidence lower bound (ELBO) and then discuss the optimization procedure of the model. During training\, the VAE learns smooth latent space representations through regularization. This learning paradigm can be applied to various tasks\, including unsupervised clustering\, regression\, and the generation of new instances. To demonstrate its application\, I will showcase how the VAE model can be applied to high-dimensional datasets and highlight the results in terms of clustering performance and sample generation.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-march-10-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20250224T140000
DTEND;TZID=UTC:20250224T150000
DTSTAMP:20260424T015823
CREATED:20250222T040452Z
LAST-MODIFIED:20250222T040711Z
UID:7476-1740405600-1740409200@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday February 24\, 2025
DESCRIPTION:Venue\, time and date: ISRT\, 2:00 pm\, February 24\, 2025 \nSpeaker: Humayera Islam\, PhD\, Postdoctoral Scholar in Precision Health at the University of Chicago \nTalk 1 \nTitle: From Statistical Models to LLMs: The Evolution of Feature Representation in Predictive Modeling \nAbstract: \nWith the digitization of healthcare and public health systems\, data collection has expanded far beyond traditional numerical and categorical formats to include complex modalities such as natural language (e.g.\, clinical notes)\, medical images (e.g.\, radiology scans)\, genetic data (e.g.\, omics)\, and temporally extensive time-series data (e.g.\, electronic health records). This expansion was driven by advancements in data storage capacity\, enabling the collection of massive\, high-dimensional datasets. As the size and complexity of data grew\, so did the need for more sophisticated feature representation techniques to effectively capture the underlying patterns for predictive tasks to enhance clinical decision making. This seminar traces the evolution of feature representation from traditional statistical models\, which relied on manual feature engineering\, to machine learning models that automated feature extraction\, to deep learning architectures that learned hierarchical and temporal features\, and finally to Large Language Models (LLMs) that leveraged self-attention mechanisms for contextual sequence modeling. The aim is to spark curiosity and inspire students to explore how to effectively handle these diverse data modalities and harness the power of advanced models for innovative research projects. \nTalk 2 \nTitle: Pathways to Growth: Preparing for Data Science and Informatics Graduate Programs in the US \nAbstract: \nThis talk offers a comprehensive roadmap for students aspiring to pursue graduate programs in data science and informatics in the US. It will cover the key skill sets essential for enhancing data science expertise\, including domain knowledge\, emerging methodologies\, technical proficiency\, and leadership in professional development. Additionally\, students will be provided with valuable resources such as open-source datasets and learning platforms to strengthen these skills during their time at ISRT and effectively prepare for graduate studies.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-february-24-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20250204T113000
DTEND;TZID=UTC:20250204T130000
DTSTAMP:20260424T015823
CREATED:20250201T174134Z
LAST-MODIFIED:20250201T174134Z
UID:7409-1738668600-1738674000@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Tuesday February 4\, 2025
DESCRIPTION:Title: Imputation-Based Q-Learning for Optimizing Dynamic Treatment Regimes with Right-Censored Survival Outcome \nVenue\, time and date: ISRT\, 11:30 pm\, February 4\, 2025 \nSpeaker: Abdus S. Wahed\, PhD\, Professor of Biostatistics\, Department of Biostatistics and Computational Biology\, University of Rochester\, USA \nAbstract: \nQ-learning has been one of the most commonly used methods for optimizing dynamic treatment regimes (DTRs) in multistage decision-making. Right-censored survival outcome poses a significant challenge to Q-Learning due to its reliance on parametric models for counterfactual estimation which are subject to misspecification and sensitive to missing covariates. In this paper\, we propose an imputation-based Q-learning (IQ-learning) where flexible nonparametric or semiparametric models are employed to estimate optimal treatment rules for each stage and then weighted hot-deck multiple imputation (MI) and direct-draw MI are used to predict optimal potential survival times. Missing data are handled using inverse probability weighting and MI\, and the nonrandom treatment assignment among the observed is accounted for using a propensity-score approach. We investigate the performance of IQ-learning via extensive simulations and show that it is more robust to model misspecification than existing Q-Learning methods\, imputes only plausible potential survival times contrary to parametric models and provides more flexibility in terms of baseline hazard shape. Using IQ-learning\, we developed an optimal DTR for leukemia treatment based on a randomized trial with observational follow-up that motivated this study. \nPaper link: \nImputation-Based Q-Learning for Optimizing Dynamic Treatment Regimes with Right-Censored Survival Outcome | Biometrics | Oxford Academic 
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-tuesday-february-4-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20250127T140000
DTEND;TZID=UTC:20250127T150000
DTSTAMP:20260424T015823
CREATED:20250125T043720Z
LAST-MODIFIED:20250125T043905Z
UID:7375-1737986400-1737990000@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, January 27\, 2025
DESCRIPTION:Title: Bridging Theory and Practice – A Journey Through Applied Statistics Across Academia\, Industry\, and Consulting (Part 2) \nVenue and time: ISRT\, 2:00 pm \nSpeaker: Yeasmin Khandakar\, PhD\, Senior Data Scientist\, Transurban Limited\, Australia \nAbstract: \nIn this presentation\, Dr. Yeasmin Khandakar will share her journey emphasizing how challenges can be transformed into opportunities. The first part of the presentation highlights the outcome achieved by navigating a steep learning curve and embracing uncertainties. illustrating how these challenges have been pivotal in her career and personal growth. Dr. Khandakar will highlight key moments from her career\, including her influential work on automatic ARIMA forecasting with Prof. Rob J. Hyndman. \nThe second part of the presentation will focus on the practical application of her current research in the industry. Yeasmin will discuss her role at Transurban\, where she leverages time series forecasting models and data analysis to address real-world problems. The presentation aims to inspire and equip students with insights and strategies for navigating their own professional paths.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-january-27-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20250120T140000
DTEND;TZID=UTC:20250120T150000
DTSTAMP:20260424T015823
CREATED:20250116T043311Z
LAST-MODIFIED:20250119T044955Z
UID:7369-1737381600-1737385200@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, January 20\, 2025
DESCRIPTION:Title: Bridging Theory and Practice – A Journey Through Applied Statistics Across Academia\, Industry\, and Consulting \nVenue and time: ISRT\, 2:00 pm \nSpeaker: Roman Ahmed\, PhD\, Senior Analytics Specialist (Statistics and Machine Learning) at Optus Australia \nAbstract: \nThis presentation traces the learning and professional journey of a statistician who has navigated a diverse career across academic\, industrial\, commercial research\, government\, and corporate consulting sectors. Beginning with a BSc in Applied Statistics and PhD in Econometrics and Business Statistics in the early 2000s\, the speaker will share key insights from their evolution as a statistician\, highlighting the ways in which their expertise in applied statistics has been shaped by each environment. From theoretical foundations in academia to practical applications in industry\, government\, and consulting\, the presentation will focus on the challenges\, strategies\, and tools employed to bridge the gap between statistical theory and real-world problem-solving. Through personal anecdotes and case studies\, the speaker will demonstrate the versatility of statistical knowledge in solving complex problems across various domains\, emphasizing the importance of adaptability\, communication\, and continuous learning in a dynamic professional landscape. This reflection on their journey offers valuable lessons for aspiring statisticians seeking to understand the breadth of opportunities and challenges that exist beyond the classroom.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminars-on-monday-january-20-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20250113T140000
DTEND;TZID=UTC:20250113T170000
DTSTAMP:20260424T015823
CREATED:20250107T065038Z
LAST-MODIFIED:20250107T065038Z
UID:7360-1736776800-1736787600@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminars on Monday\, January 13\, 2025
DESCRIPTION:Seminar 1\nTitle: Opportunities and Challenges in Single-Cell RNA-Seq: Revealing Biomarkers and Regulatory Networks in Early Brain Development \nVenue and time: ISRT\, 2:00 pm \nSpeaker: Dr. Md. Alamin\, Assistant Professor\, Department of Mathematics & Physics\, School of Engineering and Physical Sciences\, North South University \nAbstract: \nUnraveling the molecular mechanisms of early neuronal development is critical to understanding the genetic and regulatory factors driving human brain formation. Leveraging single-cell RNA sequencing (scRNA-seq)\, we profiled the transcriptional dynamics during the differentiation of human embryonic stem cells (hESCs) into neurons at two key time points: Day 26 (D26) and Day 54 (D54). Our analysis uncovered 539 differentially expressed genes (DEGs)\, revealing that up-regulated DEGs are involved in neurogenesis\, while down-regulated DEGs play roles in synapse regulation. Reactome pathway analysis highlighted significant contributions of down-regulated DEGs to synaptic protein interactions. Furthermore\, we identified 20 critical transcription factors and explored miRNA-DEG and TF-miRNA interactions\, advancing our understanding of gene regulatory networks during early brain development. These findings offer valuable insights into the genetic underpinnings of intelligence\, mental health\, and neurodevelopmental disorders. Moreover\, I will address the emerging challenges for statisticians and mathematicians in analyzing and interpreting high-dimensional single-cell data. These include handling data sparsity\, developing robust computational models\, and integrating multimodal datasets to uncover complex biological interactions. By bridging the gaps between biology\, statistics\, and mathematics\, we can push the boundaries of our understanding and foster innovation in neuroscience research. \nSeminar 2\nTitle: A reflection on the recent development of the subject area of Statistics and some specific issues of concern for further research \nVenue and time: ISRT\, 3:00 pm \nSpeaker: Dr. Moudud Alam\, Associate Professor in Microdata Analysis\, Dalarna University\, Sweden \nAbstract: \nThis talk is divided into two parts. In first part the recent development of the subject area of Statistics\, particularly the development of Data Science\, is discussed from the speaker’s experience along the way to develop a master’s and a PhD programme in Data Science\, and Data Analytics at a Swedish university. The media outcry and the popularity of the computing technology\, and artificial intelligence has pushed the Statistics community to rethink about its longstanding branding. Dedication of the 2024 Nobel prize in Physics and partly in Chemistry\, to the contribution in the development of artificial neural network\, and artificial intelligence can be considered as yet another dictation from the scientific community of the future direction of the filed. The contemporary labour market demands of the computing and soft skills is a non-negligible factor influencing the current trend of the subject. In this talk the experience of the speaker’s journey from Statistics to Data Science is discussed\, in connection with the related Swedish and European initiatives. Yet\, in this impassionate outrage it seems the scientific community is undermining (if not missing) the core assignment\, as all researchers concentrate too much to the practical applications driven by the contemporary problems\, mainly coming from the industry. In the second part\, the speaker draws attention\, using literature review and own research towards a number of statistical core issues that need special attention. In particular\, the limitation of ad hoc (such as cross validation) inferential procedure is exemplified using variable selection problem as an example\, from the literature. Using the speaker’s own research on service lifetime estimation of traffic signs in Sweden the speaker highlights the need for core statistical skills in dealing with unconventional data sources in the Data Science era. \n  \n 
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminars-on-monday-january-13-2025/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20241223T140000
DTEND;TZID=UTC:20241223T150000
DTSTAMP:20260424T015823
CREATED:20241222T010115Z
LAST-MODIFIED:20241222T011925Z
UID:7338-1734962400-1734966000@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, December 23\, 2024
DESCRIPTION:Title: Postpartum family planning counselling during maternity care visits in Bangladesh and its effect on contraceptive initiation \nVenue and time: ISRT\, 2:00 pm \nSpeaker: Md. Moinuddin Hiader\, Associate Scientist\, icddr\,b \nThe speaker will initiate the talk by briefly discussing what he and his team expect from a fresh graduate as employers and what helps in career growth in public health research. He will conclude the talk by discussing 2-3 research ideas using publicly available data.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-december-23-2024/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20241111T140000
DTEND;TZID=UTC:20241111T150000
DTSTAMP:20260424T015823
CREATED:20241104T043223Z
LAST-MODIFIED:20241104T043223Z
UID:7206-1731333600-1731337200@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, November 11\, 2024
DESCRIPTION:Title: Engineering Energy Efficiency among Residential Customers in Dhaka using Home Energy Reports (HERs) \nVenue and time: ISRT\, 2:00 pm \nSpeaker: Atonu Rabbani\, Ph.D\, Professor\, Department of Economics\, University of Dhaka \nAbstract: \nIn this study\, we estimated the potential impacts of home energy reports (HERs) on energy efficient behaviours among residential customers of Dhaka\, the capital of Bangladesh. We partnered with one of the two major retail power distribution companies. Using administrative consumption data\, we developed HERs with social feedback or descriptive norms\, where customers received comparisons between their own consumption and the averages of their neighbours. Using a randomized control trial\, we compared the electric energy consumption of a group who received “placebo” reports based only on their own consumption and without any comparison groups. Our findings suggest that the energy consumption declined by about 5 percent. However\, these impacts were short-lived and there was also suggestive evidence of possible rebound effects. Consistent with prior findings\, we also found that the impacts were larger for consumers who had higher consumption based on pre-intervention energy consumption. The point estimates suggest that households who received feedbacks based on their expenditure exhibited a larger impact compared to the households who received quantity-based feedbacks. The preliminary analyses suggest that behavioural nudges can be effective in the short run and that more intense and continuous feedback may be necessary for longer-term effects.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-november-11-2024/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20240918T100000
DTEND;TZID=UTC:20240918T113000
DTSTAMP:20260424T015823
CREATED:20240915T033016Z
LAST-MODIFIED:20240916T045507Z
UID:7018-1726653600-1726659000@isrt.ac.bd
SUMMARY:A talk by Awan Afiaz on 18 September 2024
DESCRIPTION:Title: Optimal Sandwich Variance Estimator in Penalized GEE for Nearly Separated Longitudinal Binary Data with Small Samples \nVenue and time: ISRT\, 10:00 am \nSpeaker: Awan Afiaz\, PhD candidate at the Department of Biostatistics\, University of Washington Seattle\, WA\, USA and ISRT alumnus \nAbstract: \nData separation arises in both independent and correlated binary data in biomedical studies and poses a substantial challenge that can lead to unreliable estimates and misleading inferences. This problem can occur due to a small sample size\, a rare exposure or event\, a very strong predictor or a linear combination of predictors\, high within-subject correlation (ICC)\, or any combination of these issues. Penalized generalized estimating equations (GEE) have been shown to be the superior approach for handling separation in binary longitudinal data\, along with bias-corrected sandwich variance estimators. Although the sandwich variance estimator is valid under misspecification of the working correlation structure in GEE\, it is downward biased by design for small samples and requires large samples for the asymptotic advantages to take effect. This has led to the development of several modified robust variance estimators for GEE for small samples\, which motivates finding the optimal sandwich estimator in the context of penalized GEE when there is near separation (sparsity) in the data. The current study proposed a bias-corrected sandwich variance estimator for penalized GEE and compared its performance with ten extant sandwich estimators for nearly separated data using a simulation study. To motivate the need for an optimal sandwich estimator in penalized GEE\, we demonstrated that the existing small-sample based estimators provided contradictory results when using dermatophyte-toe onychomycosis trial data. The proposed sandwich estimator does not require any additional assumptions beyond those already employed by the original sandwich estimator for GEE. We evaluated the proposed sandwich estimator by assessing the ratio of the average SEs and the empirical SD and by calculating the type-I error rates for Wald tests of the regression coefficients. Our simulation studies showed that the proposed estimator yielded nominal-level type-I error rates based on Wald tests of regression coefficients\, regardless of whether the working correlation model was correctly specified. Furthermore\, while existing approaches performed well when the number of subjects was high\, the proposed estimator achieved nominal type-I error rates with sample sizes as low as 10\, even in the most extreme scenarios. Even though all existing sandwich estimators performed better as the number of subjects increased\, exhibiting the usual asymptotic behavior of sandwich estimators\, no other estimator uniformly achieved optimal performance faster (with respect to the number of subjects and ICC) than our proposed estimator
URL:https://isrt.ac.bd/event/a-talk-by-awan-afiaz-on-18-september-2024/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20240205T140000
DTEND;TZID=UTC:20240205T153000
DTSTAMP:20260424T015823
CREATED:20240204T022109Z
LAST-MODIFIED:20240204T022109Z
UID:6480-1707141600-1707147000@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, February 5\, 2024
DESCRIPTION:Two talks (20 minutes each) \nVenue and time: ISRT\, 2:00 pm \nTalk 1 \nTopic : Data Tracker Table \nSpeaker: Nur Mohammad\, 2nd year student\, ISRT \n  \nTalk 2 \nTopic : Introduction to Typst: a modern typesetting system \nSpeaker: Md. Aminul Islam Shazid\, MS student\, ISRT
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-february-5-2024/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20240201T140000
DTEND;TZID=UTC:20240201T153000
DTSTAMP:20260424T015823
CREATED:20240129T030850Z
LAST-MODIFIED:20240129T031353Z
UID:6427-1706796000-1706801400@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Thursday\, February 1\, 2024
DESCRIPTION:Title: A Introduction to Power BI – Data visualization \nVenue and time: ISRT\, 2:00 pm \nSpeaker: Tawfique Abdur Razzaque\, Senior Data Analyst\, Business Intelligence\, Therap (BD) Ltd. \n  \nWhat is Power BI? \nMicrosoft Power BI is an interactive data visualization software product developed by Microsoft with a primary focus on business intelligence. It is part of the Microsoft Power Platform. Power BI is a collection of software services\, apps\, and connectors that work together to turn various sources of data into static and interactive data visualizations. [Wikipedia] \n.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-thursday-february-1-2024/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20240122T140000
DTEND;TZID=UTC:20240122T153000
DTSTAMP:20260424T015823
CREATED:20240118T092012Z
LAST-MODIFIED:20240118T092012Z
UID:6415-1705932000-1705937400@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Thursday\, January 22\, 2024
DESCRIPTION:Title: Impact of Covid-19 on im(mobility) of informal sector workers in Dhaka\, Bangladesh \nVenue and time: ISRT\, 2:00 pm \nSpeaker: Md. Touhidul Alam\, PhD student at Lancaster Environment Centre\, Lancaster University\, UK \nAbstract: \nComparable to many fast-growing cities of the developing world\, Dhaka city contributes greatly to the national economy (36% of GDP and 32% of national employment)\, with much of the employments being in the informal sectors. During COVID-19 pandemic\, the poorest section of this informal workforce with little or no savings became the worst victim of economic turmoil and lockdown. To escape starvation and annihilation\, many had but to return to the very villages that they had left to join Dhaka’s ever-expanding informal workforce. Curiously\, following lifting of the COVID-19 restrictions\, many of the returnees did not come back to the city immediately. This trend has the potential to reduce the supply of informal workforce to the city\, although their absorption in the rural economy has many positives for the rural development – like the availability of new skills\, ideas and technology. Taken collectively\, COVID-19 might have opened-up new possibilities for poorer people to secure higher wellbeing for themselves\, as well as to unleash a new way of doing development. While studies focusing on loss of jobs\, food insecurity and coping mechanisms of informal workers in Bangladesh are emerging\, documentation of the (im)mobility of informal workers is required for better policymaking and advancing scientific knowledge.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-thursday-january-22-2024/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20240111T140000
DTEND;TZID=UTC:20240111T153000
DTSTAMP:20260424T015823
CREATED:20240110T044931Z
LAST-MODIFIED:20240110T080942Z
UID:6407-1704981600-1704987000@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Thursday\, January 11\, 2024
DESCRIPTION:Title: Estimating Causal Effects of Socio-economic Factors on the Prevalence of Cesarean Section Delivery in Bangladesh \nVenue and time: ISRT\, 1:45 pm \nSpeaker: Dr  A. H. M. Mahbub Latif\, Professor\, ISRT\, DU \nAbstract: TBA \n 
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-thursday-january-11-2024/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20240104T140000
DTEND;TZID=UTC:20240104T153000
DTSTAMP:20260424T015823
CREATED:20240102T094327Z
LAST-MODIFIED:20240102T094327Z
UID:6392-1704376800-1704382200@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Thursday\, January 4\, 2024
DESCRIPTION:Title: PARD: Patient-Specific Abnormal Region Detection in Alzheimer’s Disease Studies \nVenue and time: ISRT\, 2:00 pm \nSpeaker: Avizit Adhikary\, PhD candidate\, Florida State University \nAbstract: \nAlzheimer’s disease (AD) is the primary cause of dementia\, leading to cognitive challenges in processing new information\, handling complex tasks\, and experiencing personality fluctuations. To better understand and treat AD\, extensive research is needed to detect abnormal brain regions in an AD patient that can facilitate providing targeted medicine and improve the treatment pathways. However\, these regions may vary among the subjects due to the heterogeneity arising from demographic factors such as age and gender. Furthermore\, brain cells within a subject have inherent spatial dependence among themselves\, and a diseased cell may affect its neighboring cells to an unknown extent. In addition\, unmeasured confounders and measurement errors can partially or entirely mask the abnormal regions. All these points make these diseased regions challenging to detect. To this end\, we propose a Patient-specific Abnormal Region Detection (PARD) algorithm to identify the heterogeneous diseased regions by solving a Bayesian latent-space variable selection problem. Using Bayesian hierarchical modeling\, we account for the heterogeneity among the subjects as a large-scale variability and incorporate the inherent spatial dependence within subjects using ising priors into the latent space. A Gibbs sampling framework is derived for efficiently estimating the model parameters and hyper-parameters. The simulation study shows the superiority of the proposed algorithm over popular unsupervised learning methods. The algorithm is further applied to the resting-state MRI brain scans of subjects collected from Alzheimer’s Disease Neuroimaging Initiative (ADNI)\, and the detected regions are validated and analyzed by cross-matching with the brain’s default mode network (DMN).
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-thursday-january-4-2024/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20231106T140000
DTEND;TZID=UTC:20231106T153000
DTSTAMP:20260424T015823
CREATED:20231105T053544Z
LAST-MODIFIED:20231105T054143Z
UID:6243-1699279200-1699284600@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, November 6\, 2023
DESCRIPTION:Title: On the Development and Validation of Risk Prediction Models for Rare Outcomes \nVenue and time: ISRT\, 2:00 pm \nPresenter:  Dr. Md. Shafiqur Rahman\, Professor of Applied Statistics\, ISRT \nAbstract: \nRisk prediction models are commonly developed in clinical research to predict patients’ future health outcomes such as death or state of illness due to disease and/or to classify patients into clinical risk groups (low\, medium\, and high). Predictions from these models are useful to make joint decisions with both patient and clinician for future courses of treatment. However\, clinicians will be reluctant to use these models unless they can trust their predictions. To maximize the prediction accuracy and clinical utility of these models\, it is essential to ensure that they are rigorously developed\, validated\, and evaluated. However\, the standard process of model development and validation faces serious problems when the outcome is rare. This talk discusses the methodological challenges and possible solutions of model development and validation for data with rare outcomes. Issues are discussed providing separate examples of predictive models for binary and survival data with rare outcomes and illustrating them using both simulated and practical data.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-november-6-2023/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20231009T140000
DTEND;TZID=UTC:20231009T153000
DTSTAMP:20260424T015823
CREATED:20231006T172907Z
LAST-MODIFIED:20231006T173606Z
UID:6089-1696860000-1696865400@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, October 9\, 2023
DESCRIPTION:Title: Joint copula-frailty approach to model clustered survival data with dependent censoring (PhD proposal) \nVenue and time: ISRT\, 2:00 pm \nPresenter:  Nasrin Sultana\, PhD student at ISRT \nAbstract: \nWith the development of modern medical technology\, in many situations naturally a fraction of patients can be cured\, or the disease free\, while the non-cured patients lengthen their survival time. Recurrence of a disease or event is a very common phenomenon in biomedical studies. The uncured patients experience the recurrence of the disease because of some unobserved random effects or heterogeneity. Dependent censoring may arise in some situations when the right censoring is completely or partly caused by an event other than the terminal event of interest\, possibly due to these random effects. We aim at developing both a joint frailty model and a joint frailty copula model for such recurrent event lifetime data considering cure fraction that captures the heterogeneity due to dependent censoring. To achieve dependent censoring\, we also need to add joint modelling of recurrent time and time to death where time to death determines the censoring mechanism. A likelihood-based technique has been developed for the proposed models. The Expectation-Maximization (EM) and Monte Carlo Expectation-Maximization (MCEM) algorithms are being developed to carry out the underlying parameter estimation and inference procedures.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-october-9-2023/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20231008T150000
DTEND;TZID=UTC:20231008T163000
DTSTAMP:20260424T015823
CREATED:20231006T175159Z
LAST-MODIFIED:20231006T175159Z
UID:6091-1696777200-1696782600@isrt.ac.bd
SUMMARY:A talk by Bangladesh Data Center Company Limited on Sunday\, October 8\, 2023
DESCRIPTION:Representatives from Bangladesh Data Center Company Limited (BDCCL)\, a tier IV national data centre\, will be giving a talk this coming Sunday\, October 8\, 2023 at 3 pm on the following topic:\n\n“Empowering Future Leaders of Data-driven Ecosystem through Innovation”\n\n\nThis talk is a part of ISRT’s initiative to organize career development programs for students and to link academia with industry. The talk focuses on the future directions\, scopes and challenges of data science in Bangladesh.\n  \n 
URL:https://isrt.ac.bd/event/a-talk-by-bangladesh-data-center-company-limited-on-sunday-october-8-2023/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230918T140000
DTEND;TZID=UTC:20230918T153000
DTSTAMP:20260424T015823
CREATED:20230914T064104Z
LAST-MODIFIED:20230914T064619Z
UID:6029-1695045600-1695051000@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, September 18\, 2023
DESCRIPTION:Title: The Generalized Variable Importance Metric: A model agnostic method to identify predictor outcome relationship \nPresenter:  \nKaviul Anam khan \nPhD  in Biostatistics candidate at the Dalla Lana School of Public Health\, University of Toronto \nAssistant Professor\, Department of Statistical Sciences\, University of Toronto \n  \nAbstract: \nThe aim my research is to define importance of predictors for black box machine learning methods\, where the prediction function can be highly non-additive and cannot be represented by statistical parameters. In this paper we defined a “Generalized Variable Importance Metric (GVIM)” using the true conditional expectation function for a continuous or a binary response variable. We further showed that the defined GVIM can be represented as a function of the Conditional Average Treatment Effect (CATE) squared for multinomial and continuous predictors. Then we propose how the metric can be estimated using any machine learning models. Finally we showed the properties of the estimator using multiple simulations. While the estimators for the GVIM are consistent\, they have small sample biases. We proposed and efficient influence function based approach under some regularity conditions to perform one step correction of the bias. This research is going to significantly impact the public and clinical health sciences\, since this opens the door for effectively using modern machine learning methods in real life applications in health sciences.
URL:https://isrt.ac.bd/event/6029/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230821T140000
DTEND;TZID=UTC:20230821T153000
DTSTAMP:20260424T015823
CREATED:20230819T013713Z
LAST-MODIFIED:20230819T013713Z
UID:5978-1692626400-1692631800@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on Monday\, August 21\, 2023
DESCRIPTION:Title: Hierarchical structural component models for pathway analysis of longitudinal categorical phenotypes \nPresenter:  \nMd. Kamruzzaman\, PhD \nAssociate Professor\, Jagannath University\, Email: kzaman1@isrt.ac.bd \n  \nAbstract: \nSeveral statistical methods for pathway analysis have been developed to test the association between pathways and phenotypes of interest. Since pathways are highly correlated\, a hierarchical structural component models (HisCoM) was developed to analyze all pathways in a single model and take into consideration their correlation. HisCoM was originally developed to analyze a single phenotype using only one measurement per individual. Later\, it was extended to analyze multiple phenotypes (HisCoM-multi) and longitudinal phenotypes (HisCoM-GEE). These methods have been used to analyze continuous\, counts\, and binary phenotypes from cross-sectional\, clustered\, and longitudinal studies. In this study\, we propose a hierarchical structural component model for pathway analysis of longitudinal categorical phenotypes (HisCoM-RCateg). HisCoM-RCateg is proposed by combining the hierarchical structural component model and generalized estimating equations for correlated categorical phenotypes. HisCoM-RCateg accounts for the biological hierarchy of all biomarkers and pathways into a single model. In the simulation\, the proposed HisCoM-RCateg appeared to have high power than other existing methods and effectively controlled type I error for longitudinal multinomial phenotypes. To demonstrate the performance\, we also applied HiscoM-RCateg to two distinct types of longitudinal omics data\, namely the metabolite dataset and the metagenome dataset.
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-monday-august-21-2023/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230807T140000
DTEND;TZID=UTC:20230807T150000
DTSTAMP:20260424T015823
CREATED:20230726T145600Z
LAST-MODIFIED:20230728T120224Z
UID:5922-1691416800-1691420400@isrt.ac.bd
SUMMARY:Applied Statistics and Data Science Seminar on August 7 on "Testing linearity in rapid microbiological method validation"
DESCRIPTION:Title: Testing linearity in rapid microbiological method validation \nAbstract: \nTesting the linearity of a measurement system (MS) is required during its validation. Guidelines clearly discuss the appropriate design and analysis in testing the linearity of an MS when the data is Gaussian distributed but not for count data. This talk addresses how linearity is tested and the corresponding optimal design for testing linearity for Poisson data. \n  \nPresenter: \nDr. Abu Manju\nAssociate Director Statistics\nCenter for Mathematical Science (CMS)\,\nOrganon\, Oss\, the Netherlands \nAssistant Professor \nWittenborg University of Applied Science\, Appeldorn\, the Netherlands
URL:https://isrt.ac.bd/event/applied-statistics-and-data-science-seminar-on-august-7-on-testing-linearity-in-rapid-microbiological-method-validation/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230724T140000
DTEND;TZID=UTC:20230724T150000
DTSTAMP:20260424T015823
CREATED:20230718T040215Z
LAST-MODIFIED:20230718T112121Z
UID:5867-1690207200-1690210800@isrt.ac.bd
SUMMARY:Seminar on “Data Science Product Development In the (AWS) Cloud” at 2 pm on July 24\, 2023
DESCRIPTION:Speaker: Sheikh Samsuzzhan Alam\, Senior Data Science Developer for Operation\, Novartis Pharma\, Czech Republic \nVenue: ISRT seminar room \nDate and time: 2 pm on Monday\, 24 July 2023 \nTitle: Data Science Product Development In the (AWS) Cloud \nAbstract: Customer-facing software products are complex in nature and usually developed by multiple teams of engineers. On the other hand\, software products build for internal business teams such as finance\, marketing\, human resources\, etc. are domain-specific and small user group focused. Providing data science solutions as software services for internal business teams with full-stack developer teams can have high development costs. Businesses might fall back or hesitate to build such products due to high maintenance and developer cost. Hence in My presentation\, I would like to showcase how a Data Scientist can wear many hats and provide a full-stack data science solution\, which is easily maintainable\, reusable\, and cost-effective at least from the development point of view due to its Cloud Native nature. In my presentation\, I would like to demonstrate a real-world use case I have implemented using AWS services such as Sagemaker\, Lambda\, API gateway\, S3\, and DynamoDB with Python SDKs to develop a Natural Language Processing (NLP) application for internal business use.
URL:https://isrt.ac.bd/event/seminar-on-data-science-product-development-in-the-aws-cloud-at-2-pm-on-july-24-2023/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230619T110000
DTEND;TZID=UTC:20230619T120000
DTSTAMP:20260424T015823
CREATED:20230618T065328Z
LAST-MODIFIED:20230618T065328Z
UID:5779-1687172400-1687176000@isrt.ac.bd
SUMMARY:Applied Statistics Seminar on "Pairwise Accelerated Failure Time Models for Infectious Disease Transmission Within and Between Households"
DESCRIPTION:Abstract: Pairwise survival analysis handles dependent happenings in infectious disease transmission data by analyzing failure times in ordered pairs of individuals. The contact interval in the pair ij is the time from the onset of infectiousness in i to infectious contact from i to j\, where an infectious contact is sufficient to infect j if he or she is susceptible. The contact interval distribution determines transmission probabilities and the infectiousness profile of infected individuals. Many important questions in infectious disease epidemiology involve the effects of covariates (e.g.\, age or vaccination status) on transmission. Here\, we generalize earlier pairwise methods in two ways: First\, we introduce an accelerated failure time model that allows the contact interval rate parameter to depend on infectiousness covariates for i\, susceptibility covariates for j\, and pairwise covariates. Second\, we show how internal infections (caused by individuals under observation) and external infections (caused environmental or community sources) can be handled simultaneously. In simulations\, we show that these methods produce valid point and interval estimates and that accounting for external infections is critical to consistent estimation. Finally\, we use these methods to analyze household surveillance data from Los Angeles County during the 2009 influenza A(H1N1) pandemic.\n\n\nSpeaker: Yushuf Sharker\, Ph.D.\, Takeda Pharmaceuticals\, USA
URL:https://isrt.ac.bd/event/applied-statistics-seminar-on-pairwise-accelerated-failure-time-models-for-infectious-disease-transmission-within-and-between-households/
LOCATION:isrt seminar room
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230612T120000
DTEND;TZID=UTC:20230612T130000
DTSTAMP:20260424T015823
CREATED:20230607T174356Z
LAST-MODIFIED:20230612T052057Z
UID:5733-1686571200-1686574800@isrt.ac.bd
SUMMARY:Seminar on Classification and Clustering for RNA-seq data with variable selection 
DESCRIPTION:Speaker: Tanbin Rahman PhD\, FDA\, USA\n\nTitle: Classification and Clustering for RNA-seq data with variable selection\n\nAbstract: Clustering and classification play an important role in identifying sub-types of complex diseases as well as building a predictive model in the field of medicine. In recent years\, lowering of cost and high accuracy has made RNA-seq widely popular which is expected to continue to grow over the next few years. One of the important features of RNA-seq data is its count data structure. While there has been a great deal of literature in both clustering and classification methods\, most of them are either heuristic or suitable for continuous data and do not directly generalize to count data.\n\nIn the first part of the presentation\, we develop a negative binomial mixture model with lasso or fused lasso gene regularization to cluster samples (small n) with high-dimensional gene features (large p). A modified EM algorithm and Bayesian information criterion are used for inference and determining tuning parameters. The method is compared with existing methods using extensive simulations and two real transcriptomic applications in rat brain and breast cancer studies. The result shows the superior performance of the proposed count data model in clustering accuracy\, feature selection\, and biological interpretation in pathways. \nIn the second part of this presentation\, we will discuss a classification model based on negative binomial distribution via generalized linear model framework with double regularization for gene and covariate sparsity to accommodate three key elements: adequate modeling of count data with overdispersion\, gene selection and adjustment for covariate effect. The proposed method is evaluated in simulations and two real applications using cervical tumor miRNA-seq data and schizophrenia post-mortem brain tissue RNA-seq data to demonstrate its superior performance in prediction accuracy and feature selection.
URL:https://isrt.ac.bd/event/classification-and-clustering-for-rna-seq-data-with-variable-selection/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230227T133000
DTEND;TZID=UTC:20230227T143000
DTSTAMP:20260424T015823
CREATED:20230225T183508Z
LAST-MODIFIED:20230225T183508Z
UID:5608-1677504600-1677508200@isrt.ac.bd
SUMMARY:Special Applied Statistics Seminar on "Health Statistics in Bangladesh"
DESCRIPTION:Prof. Dr. Syed Abdul Hamid (https://ihe.ac.bd/faculty/syedabdulhamid)\, Health Institute\, Dhaka University\, will give a talk on “Health Statistics in Bangladesh” at ISRT on February 27\, 2023\, from 1.30-2.30 pm. He is an expert on Health Statistics. This talk is being arranged on the eve of National Statistics Day to be celebrated country-wide on February 27\, 2023.
URL:https://isrt.ac.bd/event/special-applied-statistics-seminar-on-health-statistics-in-bangladesh/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230207T120000
DTEND;TZID=UTC:20230207T130000
DTSTAMP:20260424T015823
CREATED:20230203T150054Z
LAST-MODIFIED:20230205T084015Z
UID:5593-1675771200-1675774800@isrt.ac.bd
SUMMARY:Applied Statistics Seminar on Tuesday (February 7\, 2023) at 12 PM
DESCRIPTION:Abstract:\nThe exchangeability of units between treatment groups is a key and typically untestable assumption for evaluating causal intervention effects in observational studies. Standard methods assuming exchangeability can yield biased treatment effect estimates if the assumption does not hold. Existing methods evaluate the sensitivity of treatment effect estimates to non-exchangeability due to unmeasured confounders only. In practice\, non-exchangeability can occur for either unmeasured confounders or reverse causality. We propose an index of sensitivity to non-exchangeability (ISENSE) to measure the impact of non-exchangeability on treatment effect estimates. Unlike existing methods\, ISENSE does not require imposing assumptions on the types\, numbers\, and distributions of unmeasured confounders\, and it can handle both unmeasured confounders and reverse causality. ISENSE is a computationally inexpensive local sensitivity method based on a Taylor-series approximation to the non-exchangeability likelihood\, evaluated at the parameter estimates under the exchangeability assumption. One can interpret ISENSE intuitively through the unit-free “MinNE” statistic values that capture the minimum non-exchangeability needed to cause important sensitivity. We evaluate ISENSE using simulation studies and illustrate its use with an example using administrative data from British Columbia\, Canada.\n\n\nPresenter:\n\n\nMd Rashedul Hoque\nPh.D. Candidate\, SFU\nMethodologist\, Statistics Canada\nTrainee Biostatistician\, Arthritis Research Canada\nMob: +1 778 882 0689
URL:https://isrt.ac.bd/event/applied-statistics-seminar-on-tuesday-february-7-2023-at-2pm/
CATEGORIES:seminar
END:VEVENT
END:VCALENDAR