BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Institute of Applied Statistics and Data Science - ECPv6.17.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://isrt.ac.bd
X-WR-CALDESC:Events for Institute of Applied Statistics and Data Science
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:20160101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=UTC:20211218T200000
DTEND;TZID=UTC:20211218T210000
DTSTAMP:20211213T132043Z
CREATED:20211213T132043Z
LAST-MODIFIED:20211213T132043Z
UID:4922-1639857600-1639861200@isrt.ac.bd
SUMMARY:Seminar on "What it means to be an Applied Statistician -- an industry perspective"
DESCRIPTION:Abstract:\nThis question was asked many times in the past. I am sure all of you have pondered about it at some point in your academic life. In this talk\, I will explain what it truly means to be an applied statistician. Spoiler alert: statistics is inherently applied from an industry perspective. We do not differentiate between a statistician and an applied statistician. \nSo are you ready to apply your skills? What do we expect a statistician to do in the industry? Are academic institutions making you industry-ready? There is no one correct answer here. It depends on the context. In this talk\, I will share my experience in the industry that would help you understand the opportunities ahead and how you can align your focus and take steps to make the most out of it. \n  \nSpeaker: \nEnayetur Raheem\, Ph.D.\nPrincipal Data Scientist at ConcertAI\n(A DaaS AI startup in Oncology)\nUnited States of America
URL:https://isrt.ac.bd/event/seminar-on-what-it-means-to-be-an-applied-statistician-an-industry-perspective/
LOCATION:Online\, Institute of Statistical Research and Training\, University of Dhaka\, Dhaka\, Please Select\, 1000\, Bangladesh
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20211019T113000
DTEND;TZID=UTC:20211019T123000
DTSTAMP:20211018T050015Z
CREATED:20210930T173830Z
LAST-MODIFIED:20211018T050015Z
UID:4853-1634643000-1634646600@isrt.ac.bd
SUMMARY:Seminar on Improved Statistical Approach for Climate Projection over Bangladesh using Downscaling of Global Climate Model Outputs
DESCRIPTION:Title: Improved Statistical Approach for Climate Projection over Bangladesh using Downscaling of Global Climate Model Outputs \nSpeaker: Md. Bazlur Rashid \nAbstract:\nBangladesh is facing from severe impacts of climate change because of its low-lying coastal\nareas\, deforestation\, and rapid human population growth\, technological and industrial\nintervention. The climate change parameters namely\, temperature\, heavy rainfall\, sea surface\ntemperature\, frequency of floods\, cyclones and storm surges are showing significant changed at\nevery year and it has a massive impact on food production which may turn into food uncertainty\nby amplifying the environmental and socio-economic pressure. The impact of climate change on\nenvironment is immeasurable and it has large threat in our country. Appropriate strategies based\non the climate information research will reduce the vulnerability of livelihoods and\ninfrastructures to future climate change and contributes to achieve sustainability in resources.\nClimate change projection poses an unprecedented challenge for meteorology\, climatology etc.\nGlobal Climate Model (GCM) has evolved from the Atmospheric General Circulation Models\n(AGCMs) broadly used for daily\, seasonal and long term climate projection. The most widely\ndocumented application is the projection of future climate conditions under several scenarios of\nincreasing atmospheric components. Over the last few eras\, GCMs have been developed to\nmatch the present climate system and to project future climate scenarios. Despite outstanding\nprogress\, GCMs do not deliver seamless simulations of reality and cannot afford the specifics\non very small spatial scales due to imperfect scientific understanding and limitations of\navailable observations in our country. For connecting the gap between the scale of GCMs and\ncrucial resolution for practical applications\, downscaling provides climate change information\nat a suitable spatial and temporal scale from the GCM data. No downscaling for Bangladesh of\ndetail temperature and precipitation has been undertaken. Current research in Bangladesh has not\naddressed seasonal based climate projections. Extreme events especially temperature and rainfall\nalong with seasonality\, under future climate in Bangladesh represent a further research gap and\nopportunity for this research. The main object of study is to develop efficient statistical\nmethods for climate projection. The specific objectives are (i) to identify suitable model with\nbias corrections for assessing and understanding climate impacts on rainfall and temperature\nusing climate model outputs; (ii) to explore the efficiency of the bias correction statistical\ndownscaling method in addressing the model-related uncertainties involved in future climate\npredictions; (iii) to classify a suitable downscaling approach for climate model data to allow\nseasonal meteorological climate impact studies and (iv) to cross check between available\nstatistical downscaling techniques for future climate projections and scenarios generation over\nBangladesh.\nTo achieve the objectives\, this research connects the gap between large and local scale climate\nvariables\, a number of statistical downscaling methods are used. A stepwise multiple linear\nregression method is used in study. One significant motivation behind the empirical statistical\ndownscaling method applied in this research is to make use of the large scales that the models \nare able to reproduce realistically to say something about local changes. Altogether GCMs have\na minimum skillful scale which means that their separate grid-box values are not a good diagram\nof the area they represent in the actual world (because computers work with discrete numbers).\nThe procedure of common EOF analysis makes it possible to identify common spatial patterns in\nreanalysis and GCM data on a scale that is good represented by climate models.\nThis study reveals that future CO 2 emissions are expected to have severe consequences for the\nwinter season in Bangladesh in terms of significant warming in the whole country. All emission\nscenarios show an increasing mean temperature in Bangladesh\, but while RCP2.6 shows the\ntemperature plateauing mid-century\, the average increase is 2 times higher in the far future\ncompared to the near future assuming RCP4.5\, and 4 times higher assuming RCP8.5. Finally\,\nthis research demonstrates that while warming may be unavoidable\, there are still opportunities\nto limit the severity of climate change in the future.
URL:https://isrt.ac.bd/event/improved-statistical-approach-for-climate-projection-over-bangladesh-using-downscaling-of-global-climate-model-outputs/
LOCATION:Online\, Institute of Statistical Research and Training\, University of Dhaka\, Dhaka\, Please Select\, 1000\, Bangladesh
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20210320T200000
DTEND;TZID=UTC:20210320T210000
DTSTAMP:20210318T081208Z
CREATED:20210318T074357Z
LAST-MODIFIED:20210318T081208Z
UID:4680-1616270400-1616274000@isrt.ac.bd
SUMMARY:Clinical Trials and application of Statistical Modeling and Machine Learning in Biomedical Data
DESCRIPTION:Title: Clinical Trials and application of Statistical Modeling and Machine Learning in Biomedical Data \n  \nAbstract: \nClinical trials/research are conducted to examine the clinical questions of practicing physicians. It is important to design trials appropriately in advance.  A randomized\, controlled trial is the ultimate design for treatment comparisons at the final confirmatory stage. The need and impact of a proper clinical trial has comprehended during COVID-19 pandemic. Over the years\, there has been substantial advancement of clinical trial design\, conduct of study and analysis of clinical trial data. In my talk\, I will discuss my experience with the evolvement of trial design including group sequential and adaptive trials\, analysis of clinical data using frequentist and Bayesian approach\, techniques to adjust for multiplicity\, Estimand framework and application of causal inference\, advanced data visualization and use of Machine Learning and Deep Learning to build predictive models for biomedical data. \nSpeaker: Jahangir Alam\, MS\n                Data Scientist Associate Director\,\n                Novartis Pharmaceutical\, New Jersey\, USA.
URL:https://isrt.ac.bd/event/clinical-trials-and-application-of-statistical-modeling-and-machine-learning-in-biomedical-data/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20201010T193000
DTEND;TZID=UTC:20201010T210000
DTSTAMP:20201030T063523Z
CREATED:20201030T063523Z
LAST-MODIFIED:20201030T063523Z
UID:4474-1602358200-1602363600@isrt.ac.bd
SUMMARY:Opportunities and Challenges for Statisticians During Pandemics
DESCRIPTION:Title “Opportunities and Challenges for Statisticians During Pandemics” \nThe speaker: Abdus S. Wahed\, Professor of Biostatistics\, School of Public Health\, University of Pittsburgh\, USA. \nDate & time: Saturday October 10\, 2020 at 7.30 PM (Dhaka time). \n  \n—————————Abstract———————————————————————– \nCOVID 19 has dramatically changed the lifestyles of people around the globe. In the midst of pandemic\, all of us are struggling to lead a “normal” life that we have been used to: many have lost their jobs\, homes\, and so on. While the first responders\, e.g.\, police\, physicians\, nurses\, grocery vendors are keeping us afloat risking their lives to COVID\, as statisticians\, we have other challenges to overcome. In this talk\, I will informally talk about challenges and opportunities that pandemic brings to the life of a statistician\, and hope to have a fruitful discussion with colleagues and students. \n———————————————————————————————————
URL:https://isrt.ac.bd/event/opportunities-and-challenges-for-statisticians-during-pandemics/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200726T193000
DTEND;TZID=UTC:20200726T210000
DTSTAMP:20201030T063410Z
CREATED:20201030T063410Z
LAST-MODIFIED:20201030T063410Z
UID:4472-1595791800-1595797200@isrt.ac.bd
SUMMARY:R Shiny app for the beginners
DESCRIPTION:Title: “R Shiny app for the beginners” \nSpeaker: Nabil Awan\, Assistant Professor (on leave)\, ISRT\, DU and PhD candidate at the University of Pittsburgh\, USA. \nDate & Time: Sunday\, July 26\, at 7.30PM \n  \nSuumary: R Shiny app is getting increasingly popular in both industry and academia. One reason is the automation that many companies are focusing on nowadays. Almost all grants in academia have a ‘technology component’ these days that often requires creating an interactive dashboard. There are competitors like Power BI\, Tableau dashboard\, etc. but those can also be integrated with R. Moreover\, using R allows a range of statistical methods that are not readily available in other software. While there are so many free materials available online to learn the R Shiny app\, some of us might have never gotten the time and opportunity to learn it. This session will take a hands-on DIY approach and help the participants create and host their first simple R Shiny app on the spot. We will emphasize explaining the ‘structure’ of the app so that the participants are able to understand the more complex apps available online. This session will also direct the participants to resources where they can learn more.
URL:https://isrt.ac.bd/event/r-shiny-app-for-the-beginners/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200713T090000
DTEND;TZID=UTC:20200713T100000
DTSTAMP:20201030T063253Z
CREATED:20201030T063253Z
LAST-MODIFIED:20201030T063253Z
UID:4470-1594630800-1594634400@isrt.ac.bd
SUMMARY:A brief introduction to Transcriptomic data analysis
DESCRIPTION:Title: “A brief introduction to Transcriptomic data analysis” \nSpeaker: Dr. Tanbin Rahman\, Postdoctoral Fellow at the MD Anderson Cancer Center\, USA. \nDate & time: Monday\, July 13\, 2020 at 9.00AM \n———————————Summary————————————————————- \nThis presentation will briefly discuss the different types of transcriptomic datasets in the field of statistical genomics. At first\, the structure of the datasets will be discussed. The preprocessing of the datasets followed by Differential Expression (DE) analysis and pathway analysis aimed at identifying the candidate genes and functional annotation of the candidate gene sets respectively\, will be discussed. Finally\, the application of supervised/unsupervised machine learning algorithms in genomic studies will be explored.
URL:https://isrt.ac.bd/event/a-brief-introduction-to-transcriptomic-data-analysis/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200630T110000
DTEND;TZID=UTC:20200630T120000
DTSTAMP:20201030T063146Z
CREATED:20201030T063146Z
LAST-MODIFIED:20201030T063146Z
UID:4468-1593514800-1593518400@isrt.ac.bd
SUMMARY:Academic Writing Skills: Useful Tips for Beginners
DESCRIPTION:Title: ‘Academic Writing Skills: Useful Tips for Beginners’\, \nSpeaker: Prof. Tamanna Howlader\, ISRT\, University of Dhaka \n  \nDate & time: Tuesday\, June 30\, 2020 at 11.30AM.
URL:https://isrt.ac.bd/event/academic-writing-skills-useful-tips-for-beginners/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200628T090000
DTEND;TZID=UTC:20200628T100000
DTSTAMP:20201030T062927Z
CREATED:20201030T062927Z
LAST-MODIFIED:20201030T062927Z
UID:4464-1593334800-1593338400@isrt.ac.bd
SUMMARY:Introduction to Deep Neural Network using R
DESCRIPTION:Title: “Introduction to Deep Neural Network using R” \n  \nSpeaker: Tuhin Sheikh\, ISRT\, University of Dhaka \nand PhD candidate at the University of Connecticut\, USA\, \n  \nDate and Time: Sunday\, June 28\, 2020 at 9.00AM \n  \n—————————–Summary———————————————— \nThe deep neural network (DNN) modelling has been considered to be a thriving topic in recent years. The DNN can be considered as a generalization of traditional regression analysis. Considering a particular objective of predicting output\, traditional regression analysis extracts low level features based on the observed input covariates. However\, when we deal with high dimensional data and a numerous input features\, low level feature extraction often leads to low prediction accuracy. The DNN on the other hand\, has been found to effective in extracting high level feature with promising prediction accuracy. In DNN\, we assume that similar to the neural system\, the input covariates go through different neurons at different layers until it reaches the final output layer. The higher the number of middle layers\, the deeper the network is. If there is no (hidden) layers in the middle\, it generalizes to only input and output layer as in traditional regression. Like regression\, DNN requires a loss function and criterion for minimization. The key difference would be\, extraction of hidden features and connecting those to the final output prediction. \nDue to the advancement of computer algorithms and emergence of interesting data\, this research field has been found to interesting in present times. Many big companies (e.g. Google\, Facebook\, Boehringer Ingelheim\, etc.) have been practicing deep neural network modelling due to the satisfactory performance. In the past\, mostly computer scientists and engineers contributed in this field. However\, the attractive mathematical foundation behind this interesting methodology got the attention of the Statisticians recently. As Statisticians\, there are huge scopes to contribute to this emerging field through statistical innovation. In this session\, the audience can expect the discussion on background and basic mathematical foundation of DNN. Also\, I will introduce an R package “Keras”\, which can be used to work with deep neural network. At the end of the discussion on this interesting topic\, I will spend some time discussing higher study experience in USA.
URL:https://isrt.ac.bd/event/introduction-to-deep-neural-network-using-r/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200624T150000
DTEND;TZID=UTC:20200624T170000
DTSTAMP:20201030T063035Z
CREATED:20201030T063035Z
LAST-MODIFIED:20201030T063035Z
UID:4466-1593010800-1593018000@isrt.ac.bd
SUMMARY:Sample Size Determination for Survey Research
DESCRIPTION:Title: “Sample Size Determination for Survey Research” \nSpeaker: Prof. Muhammad Shuaib\, ISRT\,  University of Dhaka \nDate & time: Wednesday\, June 24\, 2020\, at 3.15 PM.
URL:https://isrt.ac.bd/event/sample-size-determination-for-survey-research/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200622T110000
DTEND;TZID=UTC:20200622T130000
DTSTAMP:20201030T062808Z
CREATED:20201030T062800Z
LAST-MODIFIED:20201030T062808Z
UID:4462-1592823600-1592830800@isrt.ac.bd
SUMMARY:Academic Writing
DESCRIPTION:Title: “Academic Writing” \nSpeaker: Professor Syed Shahadat Hossain \, ISRT\, University of Dhaka \nDate and Time: Monday\, June 22\, 2020 at 11.00AM.
URL:https://isrt.ac.bd/event/academic-writing/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200617T110000
DTEND;TZID=UTC:20200617T130000
DTSTAMP:20201030T062612Z
CREATED:20201030T062612Z
LAST-MODIFIED:20201030T062612Z
UID:4458-1592391600-1592398800@isrt.ac.bd
SUMMARY:Visualizing world-wide Covid-19 data using R\, especially ggplot2
DESCRIPTION:Title:  “Visualizing world-wide Covid-19 data using R\, especially ggplot2” \n  \nSpeaker: Professor Mahbub Latif \, ISRT\, University of Dhaka \nDate &Time: Wednesday\, June 17\, 2020 at 11.00 AM.
URL:https://isrt.ac.bd/event/visualizing-world-wide-covid-19-data-using-r-especially-ggplot2/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200203T113000
DTEND;TZID=UTC:20200203T130000
DTSTAMP:20200130T024252Z
CREATED:20200130T022629Z
LAST-MODIFIED:20200130T024252Z
UID:4138-1580729400-1580734800@isrt.ac.bd
SUMMARY:Big Data Analytics: Wisdom or Folly!
DESCRIPTION:Title: Big Data Analytics: Wisdom or Folly! \n\nTime and venue: 11:30 am on 3rd of February (Monday) 2020 in ISRT Seminar Room\n\nSpeaker: Ejaz Ahmed\, PhD Professor at Brock University\, Canada\n\nAbstract:\n\nThere are hosts of buzzwords in today’s data-centric world\, and especially in digital and print media. We encounter data in every walks of life\, and for analytically and objectively-minded people\, data is everything. However\, making sense of the data and extracting meaningful information from it may not be an easy task. We come across buzzwords such as big data\, high dimensional data\, data science\, and open data without a proper definition of such words. The rapid growth in the size and scope of data sets in a host of disciplines has created a need for innovative statistical strategies analyzing such data. For example\, many private and public agencies are using sophisticated data mining strategies and/or big data analytics to reveal patterns based on collected information. Some examples of big data that have prompted demand are digital marketing\, customer service standards\, gene expression arrays\, social network modeling\, clinical\, genetics and phenotypic data. \nThe need for novel statistical strategies to analyze such data sets is pressing. This talk focuses on the development of statistical and computational strategies for a sparse regression model in the presence of mixed signals. The existing estimation methods have often ignored contributions from weak signals. However\, in real scenario many predictors altogether provide useful information for prediction\, although the amount of such useful information in a single predictor might be modest. The search for such signals\, sometimes called networks or pathways\, is for instance an important topic for those working on personalized medicine. We discuss a new “post selection shrinkage estimation strategy” that takes into account the joint impact of both strong and weak signals to improve the prediction accuracy and opens pathways for further research in such scenarios.
URL:https://isrt.ac.bd/event/big-data-analytics-wisdom-or-folly/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20200102T150000
DTEND;TZID=UTC:20200102T160000
DTSTAMP:20191230T090712Z
CREATED:20191230T084938Z
LAST-MODIFIED:20191230T090712Z
UID:3875-1577977200-1577980800@isrt.ac.bd
SUMMARY:Is Data Science the next step for the Statisticians?
DESCRIPTION:Data Science was hot a few years ago. It is no longer a hot topic. The tremendous growth in terms of using/adopting data science\, machine learning that the scientific community and the industry have observed over the past two years is noteworthy. Good thing is\, statisticians are gradually coming out of their cocoon to experience the new landscape. In this talk\, I will discuss why and how statisticians should take the next steps to learn machine learning before they (statisticians) potentially become obsolete in the industry.
URL:https://isrt.ac.bd/event/is-data-science-the-next-step-for-the-statisticians/
LOCATION:ISRT Seminar Room (3rd floor)\, Institute of Statistical Research and Training\, University of Dhaka\, Dhaka\, Please Select\, 1000\, Bangladesh
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20191203T140000
DTEND;TZID=UTC:20191203T153000
DTSTAMP:20191124T172807Z
CREATED:20191124T172807Z
LAST-MODIFIED:20191124T172807Z
UID:3762-1575381600-1575387000@isrt.ac.bd
SUMMARY:PhD seminar talk on Tuesday\, December 3\, 2019 at 2pm
DESCRIPTION:Title: Modified Inferential Methods on Restricted Parameters in Multivariate Regression Analysis: Applications in Socio-demographic Research\n\nSpeaker: Sheikh Mohammad Sayem\, PhD researcher at ISRT\, DU\n\nAbstract:\n\nEfficient and significant empirical estimate of the multivariate regression parameters will be helpful for the policymaker to make the right decisions about sophisticated interrelated issues in the dynamic world. Since the end of the twentieth century\, statisticians are going forward to develop unique working methodology for estimating and testing restricted parameters. This study reviews existing methods and suggests modified maximum likelihood estimator\, modified multivariate t statistic and modified joint confidence interval to get efficient estimates for linear restricted parameters of multivariate regression with continuous responses. A Monte Carlo experiment is conducted to examine relative performance of the modified methods. The proposed methodology has been also applied to detect the numerical nexus among socioeconomic determinants\, food expenditure and total monthly expenditure in “Haor” areas of Bangladesh. The study reveals that logarithm form of total monthly expenditure and food expenditure as multivariate continuous responses are significantly related to total operating land\, logarithm form of family size and total monthly income considering a restriction on the parameters at different level of significance. The modified methods are found to perform significantly better than the existing methods.
URL:https://isrt.ac.bd/event/phd-seminar-talk-on-tuesday-december-3-2019-at-2pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20190827T140000
DTEND;TZID=UTC:20190827T150000
DTSTAMP:20190813T104343Z
CREATED:20190813T103955Z
LAST-MODIFIED:20190813T104343Z
UID:3597-1566914400-1566918000@isrt.ac.bd
SUMMARY:PhD seminar talk on August 27 at 2pm
DESCRIPTION:Title: Improved Statistical Approach for Climate Projection over Bangladesh using Downscaling of Global Climate Model (GCM) Outputs\n\nSpeaker: Md. Bazlur Rashid\n\nAbstract:\nGlobal Climate Model (GCM) has evolved from the Atmospheric General Circulation Models\n(AGCMs) widely used for daily\, seasonal and long term weather prediction. The most widely\nrecognized application is the projection of future climate states under various scenarios of\nincreasing atmospheric elements. Over the last few decades\, GCMs have been developed to\nemulate the present climate system and to project future climate scenarios. Despite notable\ndevelopment\, GCMs do not provide perfect simulations of reality and cannot provide the details\non very small spatial scales due to incomplete scientific understanding and limitations of\navailable observations in our country. For bridging the gap between the scale of GCMs and\nrequired resolution for practical applications\, downscaling provides climate change information\nat a suitable spatial and temporal scale from the GCM data. No downscaling for Bangladesh of\ndetail temperature\, precipitation\, wind speed\, pressure and humidity has been undertaken.\nCurrent research in Bangladesh has not addressed thermal comfort under climate change. In\naddition\, current studies for Bangladesh do not look at seasonality clearly. Extreme events\n(rainfall and temperature)\, thermal comfort\, humidity and wind characteristics\, along with\nseasonality\, under future climate in Bangladesh represent a further research gap and opportunity\nfor this research. The main object of study will attempt to find and/or develop efficient statistical\nmethods/tools for climate projection. The specific objectives are (i) to identify suitable model\nwith bias corrections for climate projection using climate model outputs; (ii) to explore the\neffectiveness of the bias correction statistical downscaling method in addressing the model-\nrelated uncertainties involved in future climate predictions; (iii) to identify a suitable\ndownscaling approach for climate model data to allow daily/ monthly or seasonal meteorological\nclimate impact studies; (iv) to develop modify algorithms for quantifying the time-variant\nuncertainty associated with meteorological extreme weather systems and persistent events under\nfuture climate scenarios and (v) to cross check between available statistical downscaling\ntechniques for future climate projections and scenarios generation over Bangladesh. So\, GCM\ninformation can be enhanced for better representation of the conditions in specific places by\nusing historically observed local climate information from weather stations. Statistical\nDownscaling of climate model can include independent components of climatic signals like El\nNino-Southern Oscillations (ENSO)\, Indian Ocean Dipole (IOD) and Pacific Decadal\nOscillations (PDO) for a long-lead weather forecast. So\, this information can also be used to\nimprove the future climate projections to assess potential impacts and guide climate-smart\ndecisions about climate resilience in Bangladesh.
URL:https://isrt.ac.bd/event/3597/
LOCATION:isrt seminar room
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20190730T140000
DTEND;TZID=UTC:20190730T150000
DTSTAMP:20190727T010829Z
CREATED:20190727T010829Z
LAST-MODIFIED:20190727T010829Z
UID:3560-1564495200-1564498800@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, 30th July 2019 from 2:00-3:00 pm
DESCRIPTION:Title: Optimum designs for multiple objectives \nSpeaker: Mahbub Latif\, PhD \nProfessor\, ISRT \nUniversity of Dhaka \nVenue: ISRT Seminar Room \nAbstract: \nTBA
URL:https://isrt.ac.bd/event/seminar-on-tuesday-30th-july-2019-from-200-300-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20190115T140000
DTEND;TZID=UTC:20190115T150000
DTSTAMP:20190110T134806Z
CREATED:20190107T065821Z
LAST-MODIFIED:20190110T134806Z
UID:2996-1547560800-1547564400@isrt.ac.bd
SUMMARY:Seminar on 15 January Tuesday at 2 pm
DESCRIPTION:Title: Long-run relationship between the unemployment rate and the trade balance in the United States: an empirical analysis \nSpeaker: Haydory Akbar Ahmed \nMissouri State University \nDepartment of Economics\, 901 S National Avenue \nSpringfield\, MO 65897\, USA \n  \nAbstract: \nDynamics between the unemployment rate and the trade balance has both economic and political relevance.\nFrom a macroeconomic perspective\, the presence of a long run co-movement or equilibrium relationship\nalong with the nature of the short-run fluctuations in the long run co-movement will help us analyze the\ndynamics between the two in an objective manner. We use quarterly data from 1947: Q1 to 2017: Q4 on\nunemployment and trade balance to GDP ratio in the United States. Although traditional cointegration\ntests fail to detect a statistically significant long run co-movement\, a couple of threshold cointegration tests\nconfirm statistical evidence in favor of threshold cointegration or threshold long run co-movement between\nthe two. Estimated threshold vector error-correction model shows statistically significant evidence of falling\nunemployment rate with deteriorating trade balance. This finding indicates that as unemployment rates go\ndown\, the trade balance deteriorates to maintain the long run co-movement. Arguably\, as unemployment\ndecline imports rise faster than exports causing the trade balance to deteriorate.
URL:https://isrt.ac.bd/event/seminar-on-15-january-tuesday-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20181021T140000
DTEND;TZID=UTC:20181021T153000
DTSTAMP:20181019T063917Z
CREATED:20181019T063917Z
LAST-MODIFIED:20181019T063917Z
UID:2834-1540130400-1540135800@isrt.ac.bd
SUMMARY:Seminar on Sunday\, 21 October 2018 at 2 pm
DESCRIPTION:Title: Joint Modeling of Longitudinal Response and Time-To-Event Data Using Conditional Distributions: A Bayesian Perspective\n\n \nSpeaker: Arindom Chakraborty\, PhD\n\n               Assistant Professor\n               Department of Statistics\n               Visva-Bharati University\, India\n \nAbstract: Over last twenty or more years a lot of methodological development and clinical application of joint models of longitudinal and time-to-event outcomes have come up. In these studies patients are followed until an event\, such as death occurs. In most of the articles\, using subject specific random effects as frailty\, the dependency of these two processes has been established. In this article\, we propose a new joint model that consists of a linear mixed effects submodel for longitudinal data and an accelerated failure model for the time-to-event data. These two submodels are linked together by not only latent random process\, but also by the conditional distributional assumption. This model will capture the dependency of the time-to-event on the longitudinal measurements more directly. Using standard priors\, a Bayesian method is developed for estimation. All computations based on the Bayesian inference via MCMC is implemented using OpenBugs. Our proposed method is evaluated by a simulation study which shows the efficiency of the conditional model over two other models: joint models with local independence and independent models. One clinically motivating data on Duchenne muscular dystrophy (DMD) syndrome and a popular data on AIDS are also analyzed.
URL:https://isrt.ac.bd/event/seminar-on-sunday-21-october-2018-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180717T140000
DTEND;TZID=UTC:20180717T150000
DTSTAMP:20180709T074018Z
CREATED:20180709T074018Z
LAST-MODIFIED:20180709T074018Z
UID:2670-1531836000-1531839600@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, July 17 at 2 pm
DESCRIPTION:Title: Non-inferiority testing under generalized Poisson distribution\n\nVenue: ISRT seminar room \n\nSpeaker:  Md Abu Manju\, PhD\n                Department of Mathematics and Computer Science\n                Eindhoven University of Technology\, Eindhoven\, The Netherlands\n\n \nAbstract: In recent years\, non-inferiority studies for count data have been increasingly used in the evaluation of new test methods or new treatments (e.g.\, microbiological test methods\, particle counters or drugs\, vaccines) and the Poisson distribution is commonly assumed as it provides a standard framework for the analysis of count data. A generalization of the Poisson distribution\, referred to as the generalized Poisson distribution (GPD)\, which models not only overdispersion\, but also underdispersion\, may be more appropriate in certain applications\, including comparison of microbiological test methods. We therefore propose tests for the assessment of non-inferiority and sample size calculation procedures under GPD. Asymptotic likelihood ratio test (LRT)\, Wald and Exact conditional tests are derived\, and the type I error rate and statistical power are computed for these three tests based on simulations. In terms of type I error rate and statistical power\, LRT and Wald test perform similarly\, but the Exact test is conservative and has less type I error rate and power than the LRT and Wald test. Finally\, expressions are derived for calculating the sample sizes that yield sufficient power to test the non-inferiority of the new methods or treatments based on Wald test and LRT.
URL:https://isrt.ac.bd/event/seminar-on-tuesday-july-17-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180606T103000
DTEND;TZID=UTC:20180606T120000
DTSTAMP:20180604T112349Z
CREATED:20180604T112349Z
LAST-MODIFIED:20180604T112349Z
UID:2568-1528281000-1528286400@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, June 6\, 2018 at 10:30 am
DESCRIPTION:Title of the Talk: Can We Train Machine Learning Methods to Outperform the High-dimensional Propensity Score Algorithm?\n\nSpeaker: Dr. M. Ehsan Karim\n\n               Assistant Professor\, SPPH\, UBC\n               Scientist & Biostatistician\, CHEOS\, Canada\n\nVenue: ISRT Seminar room\n\n—————————————Abstract-——————————————————————-\nThe use of retrospective health care claims datasets is frequently criticized for the lack of complete information on potential confounders. Utilizing patient’s health status–related information from claims datasets as surrogates or proxies for mismeasured and unobserved confounders\, the high-dimensional propensity score algorithm enables us to reduce bias. Using a previously published cohort study of postmyocardial infarction statin use (1998–2012)\, we compare the performance of the algorithm with a number of popular machine learning approaches for confounder selection in high-dimensional covariate spaces: random forest\, least absolute shrinkage and selection operator\, and elastic net. Our results suggest that\, when the data analysis is done with epidemiologic principles in mind\, machine learning methods perform as well as the high-dimensional propensity score algorithm. Using a plasmode framework that mimicked the empirical data\, we also showed that a hybrid of machine learning and high-dimensional propensity score algorithms generally perform slightly better than both in terms of mean squared error\, when a bias-based analysis is used. This talk is based on a joint work with Menglan Pang and Robert W Platt from McGill University [Epidemiology: 2018\,29(2):191–198].\n———————————————————————————————————————–
URL:https://isrt.ac.bd/event/seminar-on-wednesday-june-6-2018-at-1030-am/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180515T140000
DTEND;TZID=UTC:20180515T153000
DTSTAMP:20180515T040605Z
CREATED:20180515T040605Z
LAST-MODIFIED:20180515T040605Z
UID:2548-1526392800-1526398200@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, 15 May 2018 at 2pm
DESCRIPTION:Title: Modified Inferential Methods on Restricted Parameters in Multivariate Regression Analysis: Applications in Socio-demographic Research \nSpeaker: Sheikh Mohammad Sayem\n\n               PhD Researcher\, ISRT\n \nAbstract: Multivariate regression analysis is getting increasing attention among the socio-demographers due to the complex nature of the interdependence of the response variables. This PhD research is to carry out detailed study on restricted parameters of multivariate regression analysis with continuous\, discrete or mixed responses to seek out or develop efficient point estimation strategy and hence develop appropriate hypothesis testing procedure for hypothesis involving restricted parameters.This seminar will give a clear idea about the specific objectives of the PhD research and about the way how the objectives will be met in line of the existing literature. In this study\, a modified maximum likelihood estimation technique is primarily used for estimating restricted parameters of multivariate regression analysis with continuous responses. A Monte Carlo simulation is carried out to evaluate the performance of the estimators. The relative efficiency of the modified maximum likelihood estimator is found to be higher than the ordinary least square or maximum likelihood estimators.
URL:https://isrt.ac.bd/event/seminar-on-tuesday-15-may-2018-at-2pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180508T140000
DTEND;TZID=UTC:20180508T150000
DTSTAMP:20180507T011847Z
CREATED:20180507T011827Z
LAST-MODIFIED:20180507T011847Z
UID:2530-1525788000-1525791600@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, April 8 at 2 pm
DESCRIPTION:Title: Approximation of the Formal Bayesian Model Comparison using the Extended Conditional Predictive Ordinatecriterion\n\nSpeaker: Md. Rashedul Hoque\, Assistant Professor\, ISRT\n \nVenue and Time: ISRT seminar room\, 2 pm\n\nAbstract: The optimal method for Bayesian model comparison is the formal Bayes factor (BF)\, according to decision theory. The formal BF is computationally troublesome for more complex models. If predictive distributions under the competing models do not have a closed form\, a cross-validation idea\, called the conditional predictive ordinate (CPO) criterion can be used. In the cross-validation sense\, this is a ‘leave-out one’ approach. CPO can be calculated directly from the Monte Carlo (MC) outputs\, and the resulting Bayesian model comparison is called the pseudo-Bayes factor (PBF). We can get closer to the formal Bayesian model comparison by increasing the ‘leave-out size’\, and at ‘leave-out all’ we recover the formal BF. But\, the MC error increases with increasing ‘leave-out size’. In this study\, we examine this for linear and logistic regression models.\nOur study reveals that the Bayesian model comparison can favor a different model for PBFcompared to BF when comparing two close linear models. So\, larger ‘leave-out sizes’ are preferred which provide result close to the optimal BF. On the other hand\, MC samples based formal Bayesian model comparisons are computed with more MC error for increasing ‘leave-out sizes’; this is observed by comparing with the available closed-form results. Still\, considering a reasonable error\, we can use ‘leave-out size’ more than one instead of fixing it at one. These findings can be extended to logistic models where closed form solution is\nunavailable.
URL:https://isrt.ac.bd/event/seminar-on-tuesday-april-8-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180424T140000
DTEND;TZID=UTC:20180424T150000
DTSTAMP:20180424T115140Z
CREATED:20180422T010527Z
LAST-MODIFIED:20180424T115140Z
UID:2468-1524578400-1524582000@isrt.ac.bd
SUMMARY:Seminar on Tuesday 24 April\, 2018 at 2 PM
DESCRIPTION:Title: Ongoing Upgrading in RMG Enterprises: Preliminary Results from a Survey\n\n\nVenue and time: ISRT seminar room\, 2 PM\n\nSpeaker: Dr. Khondaker Golam Moazzem\n               Research Director\n               Centre For Policy Dialogue (CPD) \n               Bangladesh
URL:https://isrt.ac.bd/event/seminar-on-tuesday-24-april-2018-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
ATTACH;FMTTYPE=image/jpeg:https://isrt.ac.bd/wp-content/uploads/2018/04/talk_ap24_2018-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180419T080000
DTEND;TZID=UTC:20180504T170000
DTSTAMP:20180423T224937Z
CREATED:20180423T040434Z
LAST-MODIFIED:20180423T224937Z
UID:2474-1524124800-1525453200@isrt.ac.bd
SUMMARY:ISRT's Indoor Games for 2018 started on the 19th of April 2018
DESCRIPTION:The Indoor Games Tournament 2018 of ISRT started with an inauguration ceremony on Thursday\, 19th of April 2018. This year Batch 21 (the current 4th year) is organizing the event. Badminton\, table tennis\, ludo\, chess\, dart throwing\, indoor cricket are some of the many exciting games. Students\, teachers and staffs are taking part in different events. More information about the event including fixtures and rules of games can be found from the following link: \nhttps://sites.google.com/isrt.ac.bd/indoorgame2018/home
URL:https://isrt.ac.bd/event/isrts-indoor-games-for-2018-started-on-the-19th-of-april-2018/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180417T140000
DTEND;TZID=UTC:20180417T153000
DTSTAMP:20180416T172613Z
CREATED:20180416T172613Z
LAST-MODIFIED:20180416T172613Z
UID:2455-1523973600-1523979000@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, April 17 at 2 pm
DESCRIPTION:Title: Evolution of Reproductive Health in Bangladesh\n \nSpeaker: Dr. Halida Hanum Akhter \n               Faculty Member \n               Bloomberg School of Public Health \n               John Hopkins University\, Baltimore\, USA\n\nVenue: ISRT seminar room\n               \n\nAbstract: \nIn Bangladesh\, the government family planning program started in 1965 and at NGO level in around 1952. The government program was a vertical program with only IUD (intra uterine device) and high dose oral pills as a pilot program. With a demonstration project on post-partum tubal ligation in four urban clinics\, under the Family Planning Board\, the program was implemented in the whole country. In 1974\, MFSTC (Mohammadpur Fertility Services and Training Center) started as a special project of Ministry of Health and Family Planning\, Then around 1978 family planning and MCH was integrated and named MCH based family planning program. Subsequently in 1994\, ICPD (International  Conference on Population and Development) also named as Cairo conference where Reproductive health concept and definition was introduced\, vetted and cosigned by 198 countries of the world. All these countries were the cosignatory of this ‘program of action’ document and agreed to introduce reproductive health in the country program. Reproductive health is a rights based approach committing to address Life Cycle approach to women’s health issues.  Since 1994\, many international instruments\, such as ICPD POA\, ICPD +5\, ICPD +15 monitored programs nationally and globally. \nThe speaker talks about the accomplishment thus far and the unfinished agenda in reproductive health in Bangladesh.
URL:https://isrt.ac.bd/event/seminar-on-tuesday-april-17-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180410T140000
DTEND;TZID=UTC:20180410T153000
DTSTAMP:20180330T080110Z
CREATED:20180330T075816Z
LAST-MODIFIED:20180330T080110Z
UID:2413-1523368800-1523374200@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, April 10\, 2018 at 2 pm
DESCRIPTION:Title: A model for repeated binary outcomes and comparison with quasi-likelihood based approaches\n\nSpeaker: Jahida Gulshan\n                Associate Professor\, ISRT
URL:https://isrt.ac.bd/event/seminar-on-tuesday-april-10-2018-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180403T140000
DTEND;TZID=UTC:20180403T150000
DTSTAMP:20180330T080220Z
CREATED:20180318T085610Z
LAST-MODIFIED:20180330T080220Z
UID:2399-1522764000-1522767600@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, April 3 at 2 pm
DESCRIPTION:Title: The Bangladesh paradox: health gains despite poverty and income inequality  \n\n\n\n\n\nSpeaker:  Dr. Ahmed Mushtaque Raza Chowdhury \n\n                Vice Chairperson\, BRAC\n                 and\n                 Professor of Clinical Population and Family Health\n                 Joseph L. Mailman School of Public Health\n                 Columbia University\, USA
URL:https://isrt.ac.bd/event/seminar-on-tuesday-april-3-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20180109T140000
DTEND;TZID=UTC:20180109T153000
DTSTAMP:20180106T163145Z
CREATED:20180106T163145Z
LAST-MODIFIED:20180106T163145Z
UID:2236-1515506400-1515511800@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, January 9 at 2 pm
DESCRIPTION:Title: Panjer Recursion to evaluate total claim distribution recursively\n\n \nSpeaker: Dr. Abdul H. Sharif\n              Adjunct Professor\n              Mathematical Sciences Department\n              Central Connecticut State University\, USA. \n   \nAbstract: The usual method of evaluating the distribution function of the total claims incurred in a fixed period of time\, requires the computation of many convolutions of the conditional distribution of the amount of a claim given that a claim has occurred. When the expected number of claims is large\, the computation can become unwieldy even with modern large-scale electronic computers. In this presentation\, a recursive definition of the distribution of total claims is developed for a family of claim number distributions and arbitrary claim amount distributions. When the claim amount is discrete\, the recursive definition can be used to compute the distribution of total claims without the use of convolutions. This can reduce the number of required computations by several orders of magnitude for sufficiently large portfolios.\n\nFinally\, I will discuss some current research on total claim modeling. Assuming the diversified level of the audience\, the level of mathematics will be kept to probability generating function(PGF) and moment generating function (MGF).
URL:https://isrt.ac.bd/event/seminar-on-tuesday-january-9-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20171219T140000
DTEND;TZID=UTC:20171219T150000
DTSTAMP:20171105T044817Z
CREATED:20171105T044808Z
LAST-MODIFIED:20171105T044817Z
UID:2051-1513692000-1513695600@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, December 19 at 2 pm
DESCRIPTION:Title: Developing Public Speaking Skills in Statistical Communication\n \nSpeaker: Mohammed Shahidullah\, PhD\, MPH\n              State Demographer\n              Illinois Department of Public Health and\n              Adjunct Faculty\n              The University of Illinois\, Springfield\, USA\n    \n Abstract: Public speaking is the process or act of performing a speech to a live audience. The objective of public speaking is to inform\, persuade or to entertain the audience. It helps in skill building\, professional credibility\, networking and finding suitable jobs. As an applied statistician it is a challenge how we communicate with our audience including our students\, colleagues and consulting clients. My talk will concentrate on: Benefits of Public Speaking\, Steps in Successful Public Speaking and Resources Available to Improve Public Speaking. Toastmasters International club (www.toastmasters.org) is a great way to learn how to formulate\, express\, and sell your ideas and yourself with confidence. I will share with you my experience with this club and how it works. My recommendation will be to join such a club to improve your skills in public speaking and leadership. 
URL:https://isrt.ac.bd/event/seminar-on-tuesday-december-19-at-2-pm/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20171217T110000
DTEND;TZID=UTC:20171217T123000
DTSTAMP:20171206T021626Z
CREATED:20171206T021626Z
LAST-MODIFIED:20171206T021626Z
UID:2108-1513508400-1513513800@isrt.ac.bd
SUMMARY:Seminar on Sunday\, December 17 at 11 am
DESCRIPTION:Title: Testing equality of two normal means using combined samples of paired and unpaired data\n\nSpeaker: Professor Nizam Uddin \, Department of Statistics\, University of Central Florida\n \nAbstract: A test statistic for testing equality of two normal means when data consist of both paired and unpaired observations is proposed. The proposed test statistic is compared with two other standard methods known in the literature with respect to the type I error rate and power using simulation results.\n\nHe will also talk about the Big data analytics PhD program offered in the University of Central Florida. 
URL:https://isrt.ac.bd/event/seminar-on-sunday-december-17-at-11-am/
LOCATION:Please Select
CATEGORIES:seminar
END:VEVENT
END:VCALENDAR