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-WR-CALNAME:Institute of Applied Statistics and Data Science
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:20090101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=UTC:20130202T153000
DTEND;TZID=UTC:20130202T170000
DTSTAMP:20170813T202815Z
CREATED:20170813T202815Z
LAST-MODIFIED:20170813T202815Z
UID:1481-1359819000-1359824400@isrt.ac.bd
SUMMARY:Seminar on Saturday\, February 2\, 2013
DESCRIPTION:Macroeconomic Determinants and Forecasting of Stock Market Capital\n\n\nJanuary 31\, 2013 – 5:51am \n\n\n\nFull Title:\nMacroeconomic Determinants and Forecasting of Stock Market Capital: Empirical Evidence from Bangladesh\n\n\nSpeaker:\nMd. Azim Uddin\, MSc\n\n\n\nBangladesh Bank and PhD Candidate in Applied Statistics\n\n\nDate/Time:\nSaturday\, February 2\, 2013\, 3:30pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nTo diagnosis stock market functioning the present study uses time series of monthly data from November\, 2001 to April\, 2012 for Bangladesh on the macroeconomic variables\, namely\, Stock Market Capital\, Broad Money\, Consumer Price Index\, 3-Months Treasury Bill Rate\, Industrial Production Index\, Exchange Rate\, Stock Market Turnover and Standard and Poor’s 500 Index\, collected from Dhaka Stock Exchange Ltd. and Bangladesh Bank. Main objective of the planned study is to develop statistical models for the Dynamic Relationship between Macroeconomic Variables and Stock Market Capital of Bangladesh\, in addition to\, to make the Comparison among Forecasting Models. The methodology of the study is mainly based on Vector Autoregressive (VAR) model\, Johansen-Juselius cointegration test\, Granger causality test\, Vector Error Correction Model (VECM)\, Impulse Response Function (IRF)\, Forecast Error Variance Decomposition (FEVD)\, GARCH-family models and Stochastic Parameter Model (SPM). Implications of this study include the following. (i) Investors should look at the systematic risks revealed by these macroeconomic variables when structuring their portfolios and diversification strategies. (ii) Policymakers should seek to minimize macroeconomic fluctuations considering the effect of macroeconomic variables changes on the stock market when formulating economic policy.
URL:https://isrt.ac.bd/event/seminar-on-saturday-february-2-2013/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20130121T153000
DTEND;TZID=UTC:20130121T170000
DTSTAMP:20170813T202957Z
CREATED:20170813T202957Z
LAST-MODIFIED:20170813T202957Z
UID:1483-1358782200-1358787600@isrt.ac.bd
SUMMARY:Seminar on Monday\, January 21\, 2013
DESCRIPTION:Arsenic variations in shallow groundwater\n\n\nJanuary 15\, 2013 – 8:01pm \n\n\n\nFull Title:\nArsenic variations in shallow groundwater of Bangladesh explained by generalized regression modeling\n\n\nSpeaker:\nM. Shamsudduha\, PhD\n\n\n\nUniversity College London\, UK\n\n\nDate/Time:\nMonday\, January 21\, 2013\, 3:30pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nThe impact of widespread groundwater-fed irrigation and increased recharge on arsenic (As) mobilization in shallow (<50 m below ground level) groundwater remains unclear. A generalized regression model (GRM) has been constructed to describe observed spatial variations in As concentrations in shallow groundwater throughout Bangladesh. The GRM reveals that\, at the national scale\, an overall increase in groundwater-fed irrigation of 1 mm/yr (1985 to 1999) is associated with a 5% lower mean As concentration in shallow groundwater. A negative association is observed between As concentrations and net changes in groundwater recharge between periods prior to (1975 to 1980) and after (1995 to 1999) the development of groundwater-fed irrigation. In summary\, the study demonstrates that: (1) As concentrations are higher where the surface geology inhibits increases in groundwater recharge in response to irrigation; and (2) As concentrations are lower where recharge fluxes have increased in response to irrigation. This analysis indicates further where increases in recharge can flush mobile As from shallow groundwater in Bangladesh.
URL:https://isrt.ac.bd/event/seminar-on-monday-january-21-2013/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20130115T153000
DTEND;TZID=UTC:20130115T170000
DTSTAMP:20170813T203146Z
CREATED:20170813T203146Z
LAST-MODIFIED:20170813T203146Z
UID:1485-1358263800-1358269200@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, January 15\, 2013
DESCRIPTION:Item Response Theory Modelling in Analysing Rating Scale Data\n\n\nJanuary 8\, 2013 – 8:55am \n\n\n\nFull Title:\nItem Response Theory Modelling in Analysing Rating Scale Data\n\n\nSpeaker:\nAsaduzzaman Khan\, PhD\n\n\n\nThe University of Queensland and The University of New England\, Australia\n\n\nDate/Time:\nTuesday\, January 15\, 2013\, 3:30pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nRating scales are increasingly used to perform patient assessments through graded responses on ordinal level of measurements. Outcomes measured in rating scales however provide only a rank ordering of scores where differences between scores composing the scale are unequal in terms of what is being measured. Scores from rating scales do not have an underlying number line with an equal interval and are unable to represent a true linear and continuous measurement. As such\, they may not be able to comply with many assumptions of parametric statistics\, and has the potential to produce invalid statistical inferences. This paper highlights current concerns in using total raw scores in measuring outcomes in clinical settings along with their potential solutions through using Item Response Theory (IRT) modelling. Results from a simulation study and other real life examples are examined to investigate the suitability of using parametric statistics in the rating scale scores (e.g. summative scores) and their IRT transformed scores. The findings support the use of IRT-based scores when using rating scales to measure patient progress\, which in turn can help clinical decision making to optimize patient care.
URL:https://isrt.ac.bd/event/seminar-on-tuesday-january-15-2013/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20130107T153000
DTEND;TZID=UTC:20130107T170000
DTSTAMP:20170813T203337Z
CREATED:20170813T203337Z
LAST-MODIFIED:20170813T203337Z
UID:1487-1357572600-1357578000@isrt.ac.bd
SUMMARY:Seminar on Monday\, January 7\, 2013
DESCRIPTION:A Joint Modeling Approach for Multiple Ordinal Processes\n\n\nJanuary 1\, 2013 – 10:39am \n\n\n\nFull Title:\nA joint modeling approach for multiple ordinal processes via generalized estimating equations\n\n\nSpeaker:\nAbdus S. Wahed\n\n\n\nUniversity of Pittsburgh\nPittsburgh\, PA\, USA\n\n\nDate/Time:\nMonday\, January 7\, 2013\, 3:30pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nTo be announced
URL:https://isrt.ac.bd/event/seminar-on-monday-january-7-2013/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20121211T140000
DTEND;TZID=UTC:20121211T150000
DTSTAMP:20170813T203523Z
CREATED:20170813T203523Z
LAST-MODIFIED:20170813T203523Z
UID:1489-1355234400-1355238000@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, December 11\, 2012
DESCRIPTION:Estimation of slope of a regression model when both variables have measurement error\n\n\nDecember 7\, 2012 – 10:11pm \n\n\n\nFull Title:\nEstimation of slope of a regression model when both variables have measurement error\n\n\nSpeaker:\nShahjahan Khan\, PhD\n\n\n\nUniversity of Southern Queensland\, Australia\n\n\nDate/Time:\nTuesday\, December 11\, 2012\, 2:00pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nIt is well known that in the presence of equation error in the regression models any estimator based on assumed knowledge of the ratio of error variances () is biased. Although Wald’s method could deal with models that include equation error\, it lacks efficiency and subject to identifiability problem. Based on the above two methods this paper introduces a new slope estimator for regression model when both variables are subject to measurement errors and the model includes equation error. The main aim of the proposed method is to improve the efficiency of the Wald’s estimator under flexible assumption on ratio of error variance. To compare the relative efficiency of the proposed estimator with the OLS\, Wald’s and Geary’s estimators simulation studies under various assumptions are undertaken. Also included a comparison of the new estimator with the method of moments estimator when  is biased due to the presence of the equation error. The simulation results show that the proposed estimator is more consistent and efficient than the other estimators.
URL:https://isrt.ac.bd/event/seminar-on-tuesday-december-11-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20121006T120000
DTEND;TZID=UTC:20121006T130000
DTSTAMP:20170813T203718Z
CREATED:20170813T203718Z
LAST-MODIFIED:20170813T203718Z
UID:1491-1349524800-1349528400@isrt.ac.bd
SUMMARY:Seminar on Saturday\, October 6\, 2012
DESCRIPTION:Some hierarchical models for space-time data\n\n\nOctober 4\, 2012 – 6:25am \n\n\n\nFull Title:\nSome hierarchical models for space-time data\n\n\nSpeaker:\nShuvo Bakar\, PhD\n\n\n\nISRT\, University of Dhaka and CSIRO\, Australia\n\n\nDate/Time:\nSaturday\, October 6\, 2012\, 1200 noon\n\n\nVenue:\nComputer Lab (2nd Floor)\n\n\n\n  \n\nABSTRACT\nIncreasingly large volumes of space-time data are collected everywhere by mobile computing applications\, and in many of these cases temporal data are obtained by registering events\, for example telecommunication or web traffic data. Having both the spatial and temporal dimensions adds substantial complexity to data analysis and inference tasks. In this talk we discuss some hierarchical models to analyse these complex space-time data where Markov chain Monte Carlo techniques are used to make inference. We also discuss a novel method for tackling large data sets with application to daily maximum 8-hour average ground level ozone concentration data from 1997 to 2006 from a large study region in the eastern United States. Finally\, we introduce a software package “spTimer” that can model space-time data and is currently available from the R archive network.
URL:https://isrt.ac.bd/event/seminar-on-saturday-october-6-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120915T120000
DTEND;TZID=UTC:20120915T130000
DTSTAMP:20170813T203854Z
CREATED:20170813T203854Z
LAST-MODIFIED:20170813T203854Z
UID:1493-1347710400-1347714000@isrt.ac.bd
SUMMARY:Seminar on Saturday\, September 15\, 2012
DESCRIPTION:Multi-class Multi-server Accumulating Priority Queue\n\n\nSeptember 12\, 2012 – 8:49pm \n\n\n\nFull Title:\nMulti-class Multi-server Accumulating Priority Queue\n\n\nSpeaker:\nAzaz Bin Sharif\, MSc\n\n\n\nUniversity of Western Ontario\, Canada\n\n\nDate/Time:\nSaturday\, September 15\, 2012\, 12:00 noon\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nThis talk presents an accumulated priority model for multi-server queues in which waiting customers accumulate priority credit at a linear rate that depends upon their priority classification. The advantage of an accumulated priority model over a classical priority queue is that the longer a customer waits\, the greater their priority – thereby ensuring a better outcome for customers of lower priority. Numerical examples illustrating the delay impact of the accumulation rates will be presented. We hope to be able to provide broad conclusions based on the numerical experiments about the nature of priority accumulation in a multi-server environment.
URL:https://isrt.ac.bd/event/seminar-on-saturday-september-15-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120627T150000
DTEND;TZID=UTC:20120627T160000
DTSTAMP:20170813T204126Z
CREATED:20170813T204126Z
LAST-MODIFIED:20170813T204126Z
UID:1495-1340809200-1340812800@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, June 27\, 2012
DESCRIPTION:Modeling Healthcare Data Using Markov Decision Process\n\n\nJune 22\, 2012 – 6:26am \n\n\n\nFull Title:\nModeling Healthcare Data Using Markov Decision Process\n\n\nSpeaker:\nKumer Pial Das\, PhD\n\n\n\nLamar University\, Texas\, USA\n\n\nDate/Time:\nWednesday\, June 27\, 2012\, 3:00pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nA Markov decision process (MDP) is a 4-tuple (a set of states\, a set of actions\, a set of rewards\, and a transition probability function). MDPS are used to study a wide range of randomization problems. MDPs find optimal solutions to sequential and stochastic decision problems. An MDP binds previous\, current\, and future system decisions through the proper definition of system states. The use of MDPs for modeling and solving medical treatment decisions has been increased significantly in recent years. This study gives an overview of MDP models and solution techniques.We describe MDP modeling in the context of healthcare data.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-june-27-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120606T120000
DTEND;TZID=UTC:20120606T130000
DTSTAMP:20170813T204306Z
CREATED:20170813T204306Z
LAST-MODIFIED:20170813T204306Z
UID:1497-1338984000-1338987600@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, June 6\, 2012
DESCRIPTION:On the Correlation and Independence Between Sample Mean and Variance\n\n\nJune 3\, 2012 – 11:52am \n\n\n\nFull Title:\nOn the Correlation and Independence Between Sample Mean and Variance\n\n\nSpeaker:\nAnwar H Joarder\, PhD\n\n\n\nDepartment of Mathematics and Statistics\nKing Fahd University of Petroleum and Minerals\nDhahran 31261\, Saudi Arabia\n\n\nDate/Time:\nWednesday\, June 6\, 2012\, 1200 noon\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nAn alternative formula for the covariance between sample mean and variance based on any probability distribution has been derived. Interesting specials cases of uncorrelation between sample mean and variance have been identified. A direct derivation of the joint moment generating function of sample mean and variance is also presented.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-june-6-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120523T103000
DTEND;TZID=UTC:20120523T113000
DTSTAMP:20170813T204441Z
CREATED:20170813T204441Z
LAST-MODIFIED:20170813T204441Z
UID:1499-1337769000-1337772600@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, May 23\, 2012
DESCRIPTION:Macroeconomic Determinants and Forecasting of Stock Market Capital\n\n\nMay 15\, 2012 – 2:47pm \n\n\n\nFull Title:\nMacroeconomic Determinants and Forecasting of Stock Market Capital: Empirical Evidence from Bangladesh\n\n\nSpeaker:\nMd. Azim Uddin\n\n\n\nStatistics Department\, Bangladesh Bank\, Dhaka\, Bangladesh\n\n\nDate/Time:\nWednesday\, May 23\, 2012\, 10:30am\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nTo diagnosis stock market functioning the present study uses time series of monthly data from November\, 2001 to April\, 2011 for Bangladesh on the macroeconomic variables\, namely\, Stock Market Capital\, Broad Money\, Consumer Price Index\, 3-Months Treasury Bill Rate and Industrial Production Index\, collected from Dhaka Stock Exchange Ltd. and Bangladesh Bank. Main objective of the planned study is to develop statistical models for the Dynamic Relationship between Macroeconomic Variables and Stock Market Capital of Bangladesh\, in addition to\, to make the Comparison among Forecasting Models. The methodology of the study is mainly based on Vector Autoregressive (VAR) model\, Granger causality test\, Johansen-Juselius cointegration test\, Vector Error Correction Model (VECM)\, Impulse Response Function (IRF)\, Forecast Error Variance Decomposition (FEVD) and Stochastic Parameter Model (SPM). Implications of this study include the following. (i) Investors should look at the systematic risks revealed by these macroeconomic variables when structuring their portfolios and diversification strategies. (ii) Policymakers should seek to minimize macroeconomic fluctuations considering the effect of macroeconomic variables changes on the stock market when formulating economic policy.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-may-23-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120509T110000
DTEND;TZID=UTC:20120509T120000
DTSTAMP:20170813T204620Z
CREATED:20170813T204620Z
LAST-MODIFIED:20170813T204620Z
UID:1501-1336561200-1336564800@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, May 9\, 2012
DESCRIPTION:An Introduction to Scientific Writing and Referencing\n\n\nMay 9\, 2012 – 9:22am \n\n\n\nFull Title:\nAn Introduction to Scientific Writing and Referencing\n\n\nSpeaker:\nDorothy Southern\, MPH\n\n\n\nCentre for Communicable Diseases\, icddr\,b\, Dhaka\n\n\nDate/Time:\nWednesday\, May 9\, 2012\, 11:00am\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nNot available
URL:https://isrt.ac.bd/event/seminar-on-wednesday-may-9-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120418T150000
DTEND;TZID=UTC:20120418T160000
DTSTAMP:20170813T204853Z
CREATED:20170813T204853Z
LAST-MODIFIED:20170813T204853Z
UID:1503-1334761200-1334764800@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, April 18\, 2012
DESCRIPTION:History of graphical representations of statistical data\n\n\nApril 13\, 2012 – 5:47am \n\n\n\nFull Title:\nHistory of graphical representations of statistical data\n\n\nSpeaker:\nOlav Muurlink\, PhD\n\n\n\nGriffith University\, Australia\n\n\nDate/Time:\nWednesday\, April 18\, 2012\, 3:00pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nThis paper briefly reviews the history of graphical representations of research findings\, and suggests that the rapid advance of published statistics have not been matched by advances in communicating these findings to readers of scholarly publications. In addressing this widening gap between statistical procedures and the communications of results\, this paper suggests that methods developed in the mid-20th century deserve to be revisited. It suggests a procedure\, Clustered Iconographic Charts (CIX)\, that enables the simultaneous presentation of multiple variables in an intuitive manner. It situates CIX within the small but growing movement towards ‘open source’ research.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-april-18-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120407T150000
DTEND;TZID=UTC:20120407T160000
DTSTAMP:20170813T205040Z
CREATED:20170813T205040Z
LAST-MODIFIED:20170813T205040Z
UID:1505-1333810800-1333814400@isrt.ac.bd
SUMMARY:Seminar on Saturday\, April 7\, 2012
DESCRIPTION:Particulate Matter and Cardiovascular Mortality: How Risky is Breathing?\n\n\nApril 5\, 2012 – 4:55am \n\n\n\nFull Title:\nParticulate Matter and Cardiovascular Mortality: How Risky is Breathing?\n\n\nSpeaker:\nSorina Eftim\, PhD\n\n\n\nGeorge Washington University\, USA\n\n\nDate/Time:\nSaturday\, April 7\, 2012\, 3:00pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nThe talk explores statistical methods in air pollution studies with examples from Dr. Eftim’s own research on the health effects of exposure to PM2.5 and mortality in the Medicare cohort. Some other examples include transboundary air pollution from forest fires\, asthma exacerbation and traffic related air pollution\, and studies of personal exposure to air pollutants.
URL:https://isrt.ac.bd/event/seminar-on-saturday-april-7-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120225T150000
DTEND;TZID=UTC:20120225T160000
DTSTAMP:20170813T205350Z
CREATED:20170813T205350Z
LAST-MODIFIED:20170813T205350Z
UID:1509-1330182000-1330185600@isrt.ac.bd
SUMMARY:Seminar on Saturday\, February 25\, 2012
DESCRIPTION:Option Pricing and Risk Management: Analytic Approaches with GARCH-Levy Dynamics\n\n\nFebruary 20\, 2012 – 11:52am \n\n\n\nFull Title:\nOption Pricing and Risk Management: Analytic Approaches with GARCH-Levy Dynamics\n\n\nSpeaker:\nMd. Sharif Ullah Mozumder\, PhD\n\n\n\nDepartment of Mathematics\, University of Dhaka\, Bangladesh\n\n\nDate/Time:\nSaturday\, February 25\, 2012\, 3:00pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nThis work considers making some contributions to the asset pricing and financial risk management literature. First of all it offers some dynamics in the area of asset pricing which are practically implementable for pricing European style options. More precisely it considers blending GARCH type non-Markovian dynamics with Levy type Markovian innovations to offer analytic valuation of European style derivatives(at this initial stage). Revealing the mathematical underpinnings–required to replace conditional Gaussian innovations in GARCH option pricing models by innovations coming from some Levy processes (with one sided and both sided jumps)–is the main focus. The necessity for this arises from the fact that the non-normal(Levy) innovations are crucial as heteroskedasticity alone doesn’t suffice to capture the option smirk and the analytic valuation is highly expected because it makes the model practically implementable. Thus besides incorporating non-normality particular attention is paid to analytic valuation as well; though the Monte Carlo techniques can be readily applied for the proposed dynamics. However an approximation is required to uphold the analytic pricing\, especially for innovations coming from Levy processes which are not Subordinator. These dynamics are capable of overcoming many deficiencies of benchmark Black-Scholes model and can be used to price other derivatives such as Credit\, Interest rate\, Commodity\, Weather etc. The approach is built on a discrete time continuous state space and upholds the no-arbitrage principle of derivative pricing through the use of conditional Esscher transform to configure Equivalent Martingale Measure(EMM). Similar to the existing literature\, established for GARCH with normal innovations\, existence of EMM provides de-facto evidence in support of no-arbitrage argument. Besides the main focus this research has made some complementary contributions to the option pricing literature. \nSince J.P.Morgan introduced RiskMetrics in 1994\, the normal quantile based VaR has been considered as industry standard for risk management. However VaR itself has inherent inconsistencies which are exacerbated under the assumption of normality. The second part of this thesis considers two frequently referred approaches to non-normality in risk management : extreme value(EV) approach and Levy approach. The idea is to reveal the relative performance of various risk measures under full density based Levy approach and solely tail observation based EV approach. We provide empirical evidence which confirms that though purely tail based risk measures value-at-risk(VaR) and its coherent version expected shortfall(ES) are well comparable under both approaches\, entire spectrum based spectral risk measure(SRM) is misleading for EV approach. Backtesting risk measure VaR is considered under both approaches. We plan to improve the computational efficiency of\nestimation of Levy coherent risk measures through application of characteristic function based FRFT. Our ultimate goal is to see whether the conditional moment generating functions developed for GARCH-Levy models in the first part of this thesis can be adapted to the characteristic function based FRFT technique in order to estimate the risk measures in analytic fashion.
URL:https://isrt.ac.bd/event/seminar-on-saturday-february-25-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120218T150000
DTEND;TZID=UTC:20120218T160000
DTSTAMP:20170813T205618Z
CREATED:20170813T205618Z
LAST-MODIFIED:20170813T205618Z
UID:1511-1329577200-1329580800@isrt.ac.bd
SUMMARY:Seminar on Saturday\, February 18\, 2012
DESCRIPTION:An overview of clinical research: Observational studies\n\n\nFebruary 14\, 2012 – 10:15am \n\n\n\nFull Title:\nAn overview of clinical research: Observational studies\n\n\nSpeaker:\nShomoita Alam\n\n\n\nCentre for Communicable Diseases\, icddr\,b\n\n\nDate/Time:\nSaturday\, February 18\, 2012\, 3:00pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nIn clinical research\, not only understanding the association between disease and exposure is important\, but also to know whether the exposure caused the disease holds equal concern. If it can be showed that the exposure caused the outcome\, the impact of the outcome can be evaluated by intervening on the exposure. An ideal experiment is unrealistic\, given that the same individual is followed for outcome under identical conditions with and without the exposure\, by going back in time. To mimic such counterfactual\, one possible way is to design a randomized control trial (RCT) where the exposure is assigned by the investigator and followed up till the outcome occurs. In many cases it may not be a feasible option to randomly assign exposure of interest in RCT due to ethical issues or as it is cost prohibitive. The best alternative is conducting observational studies. Observational studies can be either analytical or descriptive. Analytical studies feature a comparison (control) group\, whereas descriptive studies do not. Hypotheses about causation from descriptive studies are often tested in rigorous analytical studies. Within analytical studies\, cohort studies\, case control studies and cross sectional studies are popularly used among epidemiologists based on their research objective.
URL:https://isrt.ac.bd/event/seminar-on-saturday-february-18-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120129T150000
DTEND;TZID=UTC:20120129T160000
DTSTAMP:20170813T205807Z
CREATED:20170813T205807Z
LAST-MODIFIED:20170813T205807Z
UID:1513-1327849200-1327852800@isrt.ac.bd
SUMMARY:Seminar on Sunday\, January 29\, 2012
DESCRIPTION:Statistical method for the analysis of functional connections of multiple spike trains\n\n\nJanuary 19\, 2012 – 4:43pm \n\n\n\nFull Title:\nStatistical method for the analysis of functional connections of multiple spike trains\n\n\nSpeaker:\nMohammad Shahed Masud\, PhD\n\n\n\nInstitute of Statistical Research and Training\, University of Dhaka\n\n\nDate/Time:\nSunday\, January 29\, 2012\, 3:00pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nDevelopment of Multi-Electrode Arrays (MEA) enables researchers to record multiple spike trains simultaneously from associated neurons. Simultaneously recorded multiple spike trains are used to study how groups of neurons process information and how they interact with each other. Developing a statistical method for analyzing multiple spike trains and\, in particular\, estimating the functional connection between spike trains is a challenging problem that has resulted in substantial research. In this study a statistical method\, the Cox method is presented for analyzing functional connection of simultaneously recorded multiple spike trains. This method estimates a vector of influence strengths from multiple spike trains (called reference trains) to the selected (target) spike train. Selecting another target spike train and repeating the calculation of the influence strengths from the reference spike trains enables researchers to find all functional connections among multiple spike trains. In order to study functional connection an ‘influence function’ is identified. This function recognizes the specificity of neuronal interactions and reflects the dynamics of postsynaptic potential. The Cox method has been thoroughly tested using multiple sets of data generated by the neural network model of the leaky integrate and fire neurons with a prescribed architecture of connections. The results suggest that this method is highly successful for analyzing functional connection of simultaneously recorded multiple spike trains.
URL:https://isrt.ac.bd/event/seminar-on-sunday-january-29-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20120118T150000
DTEND;TZID=UTC:20120118T160000
DTSTAMP:20170814T014454Z
CREATED:20170814T014454Z
LAST-MODIFIED:20170814T014454Z
UID:1515-1326898800-1326902400@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, January 18\, 2012
DESCRIPTION:Functional data analysis and its applications\n\n\n\nJanuary 9\, 2012 – 9:41am \n\n\n\nFull Title:\nFunctional data analysis and its application\n\n\nSpeaker:\nM. Shahid Ullah\, PhD\n\n\n\nFlinders University\, Adelaide\, Australia\n\n\nDate/Time:\nWednesday\, January 18\, 2012\, 3:00pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nThe Functional Data Analysis (FDA) approach is proposed towards modelling time series data because of recognition of the need to better analyse\, model and forecast data observed over time. This approach allows for smooth functions of age\, is robust for outlying years due to wars and epidemics\, and provides a modelling framework that is easily adapted to allow for constraints and other information. Ideas from functional data analysis\, nonparametric smoothing and robust statistics are combined to form a methodology that is widely applicable to any functional time series data observed discretely and possibly with error. The model is a generalization of the Lee–Carter (LC) model and is applied to French mortality data\, Australian fertility data and Finish injury data\, and the forecasts obtained are shown to be superior to those from the LC method and several of its variants. A new approach called Forecast REsidual Sum of Squares (FRESS) is also proposed to check the forecast accuracy. Using the specific example of Finish injury data\, FRESS demonstrated that FDA is superior over the more commonly reported ordinary least square\, Poisson and negative binomial modelling approaches in terms of forecast accuracy.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-january-18-2012/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20111228T120000
DTEND;TZID=UTC:20111228T130000
DTSTAMP:20170814T014705Z
CREATED:20170814T014705Z
LAST-MODIFIED:20170814T014705Z
UID:1517-1325073600-1325077200@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, December 28\, 2011
DESCRIPTION:Bayesian Penalized Methods for High-Dimensional Data\n\n\nDecember 16\, 2011 – 7:37am \n\n\n\nFull Title:\nBayesian Penalized Methods for High-Dimensional Data and Network Analysis\n\n\nSpeaker:\nZakaria S Khondker\n\n\n\nUniversity of North Carolina at Chapel Hill\, USA\n\n\nDate/Time:\nWednesday\, December 28\, 2011\, 12:00 noon\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nThe curse of dimensionality boils down to dealing with too many parameters than the sample size reasonably permits. When dimension is larger than the sample size the model is unidentifiable and all the parameters are not estimable. Even when the dimension is smaller than the sample size but dimension to sample size ratio is not small enough or there is colinearity among the predictors the estimators are unstable. Penalized methods for shrinkage of parameters are becoming increasingly popular. The advent of high-dimensional data\, where the number of covariates (p) or responses (d) exceed the sample size (n)\, made traditional estimation techniques infeasible. Even for cases with sample size larger than the number of parameters shrinkage can improve performance in both mean and covariance\nparameters. \nOur first paper focuses on estimation of sparse covariance matrices and their inverse subject to positive definiteness constraints. The abundance of high-dimensional data\, where the sample size (n) is less than the dimension (d)\, requires shrinkage estimation methods since the maximum likelihood estimator is not positive definite in this case. Furthermore\, when n is larger than d but not sufficiently larger\, shrinkage estimation is more stable than maximum likelihood as it reduces the condition number of the precision matrix. Frequentist methods have utilized penalized likelihood methods\, whereas Bayesian approaches rely on matrix decompositions\, Wishart priors or graph theory for shrinkage. In this paper we propose a new Bayesian method\, called the Bayesian Covariance Lasso (BCLASSO)\, for the shrinkage estimation of a precision (covariance) matrix. We consider a class of priors for the precision matrix that leads to the popular frequentist penalties as special cases\, develop a Bayes estimator for the precision matrix\, and propose an efficient sampling scheme that does not precalculate boundaries for positive definiteness. The proposed method is permutation invariant and performs shrinkage and estimation simultaneously for non-full rank data. Simulations show that the proposed BCLASSO performs similarly as frequentist methods for non-full rank data. \nOur second paper focuses on estimation of the matrix of regression coefficients for high-dimensional multivariate response. The common approaches for dimension reduction in high-dimensional data include variable selection and penalized regression. Penalized approaches like lasso\, adaptive lasso\, SCAD\, and Bayesian lasso have been used for the estimation of mean parameters for multivariate response. A less explored approach for multivariate response involves dimension reduction via reduced rank decomposition of the regression coefficient matrix to take advantage of correlations among the regression coefficients that arises due to correlation among both responses and predictors. The approach may be advantageous when genes work in unison affecting each other and small effects of many genes may add up to a larger phonotypical impact. We first derive the framework for L1 priors on multivariate coefficient matrix in traditional approach. Then we develop the generalized low rank regression (GLRR) model under L2 priors and derive the framework for L1 priors. Simulations and application to ADNI data suggest that GLRR has great advantage over traditional approaches. It greatly reduces the number of parameters while performing much better; comparative performance gets even better for higher dimensions.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-december-28-2011/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20111126T143000
DTEND;TZID=UTC:20111126T153000
DTSTAMP:20170814T014913Z
CREATED:20170814T014913Z
LAST-MODIFIED:20170814T014913Z
UID:1519-1322317800-1322321400@isrt.ac.bd
SUMMARY:Seminar on Saturday\, November 26\, 2011
DESCRIPTION:Multi-stratum and split-plot designs in Industrial experiments\n\n\nNovember 19\, 2011 – 7:49pm \n\n\n\nFull Title:\nMulti-stratum and split-plot designs in Industrial experiments\n\n\nSpeaker:\nM. Lutfor Rahman\n\n\n\nSchool of Mathematical Sciences\, Queen Mary\, University of London\n\n\nDate/Time:\nSaturday\, November 26\, 2011\, 2:30 PM\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nHard-to-set factors lead to split-plot type designs and mixed models. Mixed models are used to analyze multi-stratum designs as each stratum may have random effects on the responses. It is usual to use residual maximum likelihood (REML) to estimate random effects and generalized least squares (GLS) to estimate fixed effects. But a typical property of REML-GLS estimation is that it gives highly undesirable and misleading conclusions in non-orthogonal split-plot designs with few main plots. To overcome the problem a Bayesian method considering informative priors for variance components and using Markov chain Monte Carlo (MCMC) sampling would be an alternative approach. In the current study we have implemented MCMC techniques in two industrial experiments. Mixed binary logit and mixed cumulative logit models were considered for binary and categorical responses respectively. Deviance information criterion (DIC) was used to choose the best models in different scenarios.
URL:https://isrt.ac.bd/event/seminar-on-saturday-november-26-2011/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20110209T150000
DTEND;TZID=UTC:20110209T160000
DTSTAMP:20170814T015102Z
CREATED:20170814T015102Z
LAST-MODIFIED:20170814T015102Z
UID:1521-1297263600-1297267200@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, February 9\, 2011
DESCRIPTION:Robustness of testing correlation and equality of variances\n\n\nFebruary 9\, 2011 – 4:41am \n\n\n\nFull Title:\nRobustness of testing correlation and equality of variances\n\n\nSpeaker:\nAnwar H Joarder\, PhD\n\n\n\nKing Fahd University of Petroleum & Minerals\nDhahran\, Saudi Arabia\n\n\nDate/Time:\nWednesday\, February 9\, 2011\, 1500\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nMany statistical methods say testing correlation and equality of variances\, have been developed under the assumptions of independence\, identicality and normality of observations. In this talk\, we will address how far we can relax the three strong assumptions with special reference to distributions that shares many intrinsic properties of normal distribution but distinct in many respects. These are bivariate t-distribution\, bivariate contaminated normal distribution and many other distributions belonging to bivariate elliptical distributions.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-february-9-2011/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20110205T150000
DTEND;TZID=UTC:20110205T160000
DTSTAMP:20170814T015235Z
CREATED:20170814T015235Z
LAST-MODIFIED:20170814T015235Z
UID:1523-1296918000-1296921600@isrt.ac.bd
SUMMARY:Seminar on Saturday\, February 5\, 2011
DESCRIPTION:Risk Prediction Models for Assessing 30-Day & Long Term Mortality\n\n\nFebruary 9\, 2011 – 4:39am \n\n\n\nFull Title:\nRisk Prediction Models for Assessing 30-Day & Long Term Mortality Following CABG/AVR in Australian Cohort\n\n\nSpeaker:\nBaki Billah\, PhD\n\n\n\nDepartment of Epidemiology and Preventive Medicine\nMonash University\, Australia\n\n\nDate/Time:\nSaturday\, February 5\, 2011\, 1500\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nNot available
URL:https://isrt.ac.bd/event/seminar-on-saturday-february-5-2011/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20110118T150000
DTEND;TZID=UTC:20110118T160000
DTSTAMP:20170814T015411Z
CREATED:20170814T015411Z
LAST-MODIFIED:20170814T015411Z
UID:1525-1295362800-1295366400@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, January 18\, 2011
DESCRIPTION:Visual statistical inference for regression parameters\n\n\nJanuary 15\, 2011 – 6:28am \n\n\n\nFull Title:\nVisual statistical inference for regression parameters\n\n\nSpeaker:\nMahbubul Majumder\, MSc\n\n\n\nDepartment of Statistics\, Iowa State University\, USA\n\n\nDate/Time:\nTuesday\, January 18\, 2011\, 3:00 PM\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nStatistical graphics play a crucial role in exploratory data analysis\, model checking and diagnosis. Until recently there were no formal visual methods in place for determining statistical significance of findings. This changed\, when Buja et al. (2009) conceptually introduced two protocols for formal tests of visual findings. In this paper we take this a step further by comparing the lineup protocol (Buja et al. 2009) against classical statistical testing of the significance of regression model parameters. A human subjects experiment is conducted using simulated data to provide controlled conditions. Results suggest that the lineup protocol provides results equivalent to the uniformly most powerful (UMP) test and for some scenarios yields better power than the UMP test.
URL:https://isrt.ac.bd/event/seminar-on-tuesday-january-18-2011/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20101223T123000
DTEND;TZID=UTC:20101223T133000
DTSTAMP:20170814T020253Z
CREATED:20170814T020253Z
LAST-MODIFIED:20170814T020253Z
UID:1533-1293107400-1293111000@isrt.ac.bd
SUMMARY:Seminar on Thursday\, December 23\, 2010
DESCRIPTION:Application of propensity scores in epidemiology\n\n\nDecember 21\, 2010 – 11:21pm \n\n\n\nFull Title:\nApplication of propensity scores in epidemiology\n\n\nSpeaker:\nEhsan Karim\, MSc\n\n\n\nDepartment of Statistics\nUniversity of British Columbia\nVancouver\, BC\, Canada\n\n\nDate/Time:\nThursday\, December 23\, 2010\, 1215\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nRandomization is recognized as a tool for removing bias in any experiment. In particular\, in Epidemiologic treatment group comparisons\, this tool is considered as a gold standard. However\, in a practical scenario\, at times\, the experiments may be poorly designed\, or may not be executed properly\, and therefore\, bias may creep in. Also\, having access to huge amount of non-randomized observational hostital data forces us to develop statistical framework of utilizing non-randomized data. Various approaches had been suggested in the literature to ensure comparability of such treated and untreated groups. The approach of propensity scores is one way to adjust for these bias sources in absense of randomization. In this talk\, I will discuss briefly the main concepts of how propensity scores are utilized\, implemented (in theory and in R/Stata/SAS) and interpreted in these type of situations\, with special emphasis on Epidemiologic data.
URL:https://isrt.ac.bd/event/seminar-on-thursday-december-23-2010-4/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20101223T113000
DTEND;TZID=UTC:20101223T130000
DTSTAMP:20170814T020055Z
CREATED:20170814T020055Z
LAST-MODIFIED:20170814T020055Z
UID:1531-1293103800-1293109200@isrt.ac.bd
SUMMARY:Seminar on Thursday\, December 23\, 2010
DESCRIPTION:Variable selection technique for AFT models\n\n\nDecember 21\, 2010 – 11:19pm \n\n\n\nFull Title:\nVariable selection technique for AFT models with regularized weighted least squares when p > n\n\n\nSpeaker:\nMd Hasinur Rahaman Khan\, MSc\n\n\n\nDepartment of Statistics\nUniversity of Warwick\nCoventry\, UK\n\n\nDate/Time:\nThursday\, December 23\, 2010\, 1130\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nVariable selection is fundamental to high-dimensional survival modeling. The regularized least squares method with appropriate penalty is a widely used method for simultaneous variable selection and coefficient estimation in linear regression\, including accelerated failure time (AFT) model in survival analysis. I will discuss a variable selection technique which is based on a regularized weighted least squares method. Quadratic programming is used to solve this regularized weighted least squares objective function with  constraints and constraints that come due to the censored observations from the right censored data. This technique can be applied to survival data with  e.g. microarray data set. Simulation studies and real data example will be provided for illustration. In the beginning of this talk I will discuss the censoring effect on the Kaplan-Meier estimates when\ndata are generated from different skewed failure time distributions.
URL:https://isrt.ac.bd/event/seminar-on-thursday-december-23-2010-3/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20101223T100000
DTEND;TZID=UTC:20101223T233000
DTSTAMP:20170814T015858Z
CREATED:20170814T015858Z
LAST-MODIFIED:20170814T015858Z
UID:1529-1293098400-1293147000@isrt.ac.bd
SUMMARY:Seminar on Thursday\, December 23\, 2010
DESCRIPTION:Simultaneous confidence intervals for multinomial proportions\n\n\nDecember 21\, 2010 – 11:16pm \n\n\n\nFull Title:\nSimultaneous confidence intervals for multinomial proportions\, their differences and ratios\n\n\nSpeaker:\nAzaz Bin Sharif\, MSc\n\n\n\nDepartment of Epidemiology & Biostatistics\nUniversity of Western Ontario\nLondon\, ON\, Canada\n\n\nDate/Time:\nThursday\, December 23\, 2010\, 1015\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nEstimation of simultaneous confidence intervals for multinomial proportions\, their differences and ratios is a long-standing problem in the literature. Existing methods suffer either from enforced symmetry and/or conservative coverage. We propose calculating confidence limits using the Wilson method or Jeffreys procedure for the multinomial proportions and for the ratios of multinomial proportions with critical values obtained from multivariate normal distributions that accounts for correlations between contrasted hypotheses. Confidence intervals for the differences are then obtained by recovering variance estimates from limits for single proportions. Simulation study shows that proposed methods perform well for a variety of parameter combinations. Data on mutational damage in Saccharomyces cerevisiae is used to illustrate the procedures.
URL:https://isrt.ac.bd/event/seminar-on-thursday-december-23-2010-2/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20101223T093000
DTEND;TZID=UTC:20101223T103000
DTSTAMP:20170814T015710Z
CREATED:20170814T015710Z
LAST-MODIFIED:20170814T015710Z
UID:1527-1293096600-1293100200@isrt.ac.bd
SUMMARY:Seminar on Thursday\, December 23\, 2010
DESCRIPTION:Mortality modeling of white spruce and black spruce\n\n\nDecember 21\, 2010 – 11:13pm \n\n\n\nFull Title:\nMortality modeling of white spruce and black spruce using combine estimators\n\n\nSpeaker:\nSuborna Ahmed\, MSc\n\n\n\nDepartment of Forest Resources Management\nUniversity of British Columbia\nVancouver\, BC\, Canada\n\n\nDate/Time:\nThursday\, December 23\, 2010\, 0930\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nUsing combined estimator approaches regular mortality models for white spruce and black spruce across the Boreal Forest of Canada are developed. For irregular time intervals between measures a modified logistic regression was used to model the annual probability of survival where permanent sample plot (PSPs) data from Alberta used. The PSP data were divided into two regions\nto represent local scale populations\, whereas the entire dataset represented the population at the larger scale. Monte Carlo simulations were used to: i) randomly sample from each region (10% and 30% sampling intensity); ii) fit the survival model by region using the sample data; iii) use each of the\ncombined estimators to weight the regional estimated parameters and obtain a larger scale model; and iv) calculate the bias and precision for each large scale model parameter and combined estimator. Therefore\, in this research\, several combined estimators were used to combine local scale models to\nobtain a general model for the regional scale to improve model precision over naive approaches for the target species.
URL:https://isrt.ac.bd/event/seminar-on-thursday-december-23-2010/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20101208T120000
DTEND;TZID=UTC:20101208T130000
DTSTAMP:20170814T020642Z
CREATED:20170814T020642Z
LAST-MODIFIED:20170814T020642Z
UID:1535-1291809600-1291813200@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, December 8\, 2010
DESCRIPTION:Public Hurricane Loss Evaluation Models: Predicting Losses of Residential Structures in the State of Florida\n\n\nDecember 21\, 2010 – 11:09pm \n\n\n\nFull Title:\nPublic Hurricane Loss Evaluation Models: Predicting Losses of Residential Structures in the State of Florida\n\n\nSpeaker:\nB. M. Golam Kibria\, PhD\n\n\n\nDepartment of Mathematics and Statistics\nFlorida International University\nMiami\, FL 33199\, USA\n\n\nDate/Time:\nWednesday\, December 8\, 2010\, 1150\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nAs an environmental phenomenon\, hurricanes cause significant property damage and loss of life in coastal areas almost every year. Although a number of commercial loss projection models have been developed to predict the property losses\, only a handful of studies are available in the public domain to predict damage for hurricane prone areas. The State of Florida has developed an open\, public model for the purpose of probabilistic assessment of risk to insured residential property associated with wind damage from hurricanes. The model comprises of the atmospheric science\, engineering and actuarial components. The atmospheric component includes modeling the track and intensity life cycle of each simulated hurricane within the Florida threat area. Based on historical hurricane statistics\, thousands of storms are\nsimulated allowing determination of the wind risk for all residential zip code locations in Florida. The wind risk information is then provided to the engineering and actuarial components to model damage and average annual loss\, respectively. The actuarial team finds the county wise loss and the total loss for the entire state of Florida. The computer center compiles all information from atmospheric science\, engineering and actuarial components\, processes all hurricane related data and completes the project. The model was submitted to the Florida Commission on Hurricane Loss Projection Methodology for approval and went through a rigorous review and was revised per the suggestions of the commission . The final model was was approved for use by the insurance companies in Florida by the commission on August 17\, 2007.\nAt every stage of the process\, statistical procedures were used to model various parameters and validate the model. This paper presents a brief summary of the main components of the model (meteorology\, vulnerability and actuarial) and then focuses on the statistical validation of the same.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-december-8-2010/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20100727T150000
DTEND;TZID=UTC:20100727T160000
DTSTAMP:20170814T020807Z
CREATED:20170814T020807Z
LAST-MODIFIED:20170814T020807Z
UID:1537-1280242800-1280246400@isrt.ac.bd
SUMMARY:Seminar on Tuesday\, July 27\, 2010
DESCRIPTION:Generalized linear models with crossed nonparametric random effects\n\n\nJuly 25\, 2010 – 4:55am \n\n\n\nFull Title:\nGeneralized linear models with crossed nonparametric random effects\n\n\nSpeaker:\nM Tariqul Hasan\, PhD\n\n\n\nDepartment of Mathematics and Statistics\nUniversity of New Brunswick\nFredericton\, NB\, Canada\n\n\nDate/Time:\nTuesday\, July 27\, 2010\, 1500\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nIn social\, health and environmental sciences\, many data are collected with a known but complex underlying structure. Over the past three decades\, multilevel modeling techniques have been extensively used to account nested data structures. However the underlying data structures are often more complex and cannot be fitted into a nested structure. This complexity of the data structure may arise due to fact that the observations are cross-classified and an observation does not belong to one member of the classification. As the existing frequentist modeling approaches have limitations to analyze such data\, we propose generalized linear models with crossed nonparametric random effect. For estimating the model parameters\, we propose the estimating equation based on BLUP (best linear unbiased predictors) of random effects. We illustrate the methodology with analysis of the child respiratory illness data.
URL:https://isrt.ac.bd/event/seminar-on-tuesday-july-27-2010/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20100310T150000
DTEND;TZID=UTC:20100310T160000
DTSTAMP:20170814T020951Z
CREATED:20170814T020951Z
LAST-MODIFIED:20170814T020951Z
UID:1539-1268233200-1268236800@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, March 10\, 2010
DESCRIPTION:Modified weights based generalized quasilikelihood inferences\n\n\nMarch 4\, 2010 – 12:08pm \n\n\n\nFull Title:\nModified weights based generalized quasilikelihood inferences in incomplete longitudinal binary models\n\n\nSpeaker:\nTaslim S. Mallick\, PhD\n\n\n\nDepartment of Statistics\, Biostatistics & Informatics\nUniversity of Dhaka\, Bangladesh\n\n\nDate/Time:\nWednesday\, March 10\, 2010\, 3:00 pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nIn an incomplete longitudinal set up\, a small number of repeated responses subject to an appropriate missing mechanism along with a set of covariates are collected from a large number of independent individuals over a small period of time. In this set up\, the regression effects of the covariates are routinely estimated by solving certain inverse weights based generalized estimating equations. These inverse weights are introduced to make the estimating equation unbiased so that a consistent estimate of the regression parameter vector may be obtained. In the existing studies\, these weights are in general formulated conditional on the past responses. Since the past responses follow a correlation structure\, the present study reveals that if the longitudinal data subject to missing mechanism are generated by accommodating the longitudinal correlation structure\, the conditional weights based on past correlated responses may yield biased and hence inconsistent regression estimates. The biasness appears to get larger as the correlation increases. As a remedy\, in this study we propose a modification to the formulation of the existing weights so that weights are not affected directly or indirectly by the correlations. We then exploit these modified weights to form a weighted generalized quasilikelihood estimating equation that yields unbiased and hence consistent estimates for the regression effects irrespective of the magnitude of correlation. The efficiencies of the regression estimates follow due to the use of the true correlation structure as a separate longitudinal weight matrix in the estimating equation.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-march-10-2010/
CATEGORIES:seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20100217T150000
DTEND;TZID=UTC:20100217T160000
DTSTAMP:20170814T021125Z
CREATED:20170814T021125Z
LAST-MODIFIED:20170814T021125Z
UID:1541-1266418800-1266422400@isrt.ac.bd
SUMMARY:Seminar on Wednesday\, February 17\, 2010
DESCRIPTION:Capacity Planning of a Perinatal Network: A Loss Network Framework\n\n\nFebruary 14\, 2010 – 3:01pm \n\n\n\nFull Title:\nCapacity Planning of a Perinatal Network: A Loss Network Framework\n\n\nSpeaker:\nMd Asaduzzaman\n\n\n\nInstitute of Statistical Research and Training\, University of Dhaka\, Bangladesh and\nUniversity of Westminster\, United Kingdom\n\n\nDate/Time:\nWednesday\, February 17\, 2010\, 3:00 pm\n\n\nVenue:\nISRT Seminar Room\n\n\n\n  \n\nABSTRACT\nNeonatal care in the UK has been organised through ’Managed clinical networks’ (MCN’s) since 2004 so that service improves and care can be provided locally. Yet over a six month period in 2006/07\, neonatal units were shut to new admissions for an average of 24 days\, and one in ten units exceeded its capacity for intensive care for more than 50 days. The North Central London Perinatal Network (NCLPN)\, one of the 5 networks in London\, consists of six hospitals\, including UCLH. The pressure on neonatal units in the NCLPN is growing very fast\, in particular at UCLH\, where the rejection of admission of neonates from neonatal intensive care unit (NICU) and high dependency unit (HDU) is now very high due to capacity shortage. We apply a loss network model with overflow based on a continuous time Markov chain for capacity planning problem of a perinatal network\, with specific application to the NCLPN. We derive expressions for overflow and rejection probabilities for each neonatal unit of the network. Results obtained with the model are very close to those observed for the UCLH. They also show that a substantial number of intensive care cots are required to keep rejection level low. Using the model\, decisions on number of cots can be made for specific levels of admission rejection probabilities for each level of care at each neonatal unit of the network and specific levels of overflow to temporary care.
URL:https://isrt.ac.bd/event/seminar-on-wednesday-february-17-2010/
CATEGORIES:seminar
END:VEVENT
END:VCALENDAR