Seminar on Stochastic Models

Tuesdays 14:15-16:00 (Krista Fischer)

Join Zoom Meetings:




25.05.21 Kristo Visk Generalized AUC as a measure to assess the discriminatory power of internal ratings-based Loss Given Default models
23.03.21 Chiara Bano

Latent Markov models for the analysis of the Russia Longitudinal Monitoring Survey

The work describes an application of a class of models for longitudinal data, called Latent Markov models, to the Russia Longitudinal Monitoring Survey. Latent Markov (LM) models are an important class of latent variable models aimed for the study of this type of data. The dataset chosen to be analysed by applying the LM models is the Russia Longitudinal Monitoring Survey, which is a panel survey, obtained from face-to-face interviews and surveyed annually since 1992. The aim of this work is to detect different groups of individuals having different levels of satisfaction and how these degrees of satisfaction evolve across years, depending on the socio-demographic and economical individual covariates.

03.11.20 Krista Fischer The (controversial) concept of biological age
27.10.20 Kaur Lumiste

I will shortly show you the possibilities of Shiny package in R (a package that enables you to make interactive dashboards very easily). I will show how to make you Shiny applications available on server or server.

Some examples of Shiny dashboards:
- a students' Shiny app on videogame data (see more on course "Andmeteadus ja visualiseerimine"  - Caution! Not all might work. More on this in the seminar).

20.10.20   World Statistics Day 2020
13.10.20 Tõnu Kollo

Tõnu Kollo will give an overview of the paper written jointly with Marju Valge about testing basic covariance structures for the t-distributed population.

Expressions of test-statistics for  the likelihood ratio test, Rao's
score test and Wald's score test have been derived, also approximations of these statistics are given.
15.09.20 Mohammad Jamsher Ali Probability of up-crossing before ruin for a Levy process having two sided jumps
10.03.20 Märt Möls Censoring and truncation - what they are; why to use; how to use?
25.02.20 Jüri Lember In seminar today, we continue to study the very basic of Bayesian statistics. In the last seminar we saw that putting priors cmpletely changes the model even in the most simplest case.
The practitioners typically do not realize that, and so the outcome of Bayesian statistical analysis might be largely unexpected (because the mudel has been changed). This obviously raises an important question -- is there any point of doing  Bayesian statistics at all? To study that question, we performed a real data analysis on protein alignment data.
18.02.20 Jüri Lember Big Bayesian Illusion. In seminar I shall speak about (mis)understanding Bayesian. It is an tutorial introduction to latent variable (or hierarchical) Bayesian models on most basic level that is meant for everyone and entitled as Big Bayesian Illusion.
10.12.19 Annika Krutto  
03.12.19 Raivo Kolde Epidemioloogia elektroonilistel terviseandmetel
19.11.19 Kaur Lumiste Automated homework testing script in R
29.10.19 Tõnu Kollo

About European Day of Statistics and its celebration in Paris last week and an overview about Holger Rootzen's talk about extreme value distributions.

15.10.19 Merli Mändul Genetic score for any secondary cardiovascular event and its predictive ability
08.10.19 Krista Fischer Challenges in biobank-based survival analysis
24.09.19   Suvised konverentsid / memories of conferences
04.06.19 Artur Sepp, PhD (Quantica Capital AG, Zürich) Trend following strategies.
  • Quantitative methods for implementing TF strategies
  • The convexity profile of TF and other systematic startegies
  • New portfolio theory incorporating tail risk
30.04.19 Mohammad Jamsher Ali

Ruin probability for merged risk processes with correlated arrivals

16.04.19 Kaur Lumiste R-i MOOCi kursuse korraldamisest
09.04.19   Tartu mitmemõõtmelise statistika konverents 2020, koosolek

Kaur Alasoo (LTAT)

Ravel Riik (MS magistrant)

“Quantifying variance explained by gene-environment interactions in molecular data"/ "Geeni-keskkonna koosmõjude poolt kirjeldatud varieeruvuse hindamine molekulaarsetes andmetes"

"Kodulaenude portfelli riskide korrigeermisest Kalmani filtri abil"

26.03.19 Kristi Läll valmiva doktoritöö "Risk scores and their predictive ability for common complex diseases" tutvustamine. Juhendaja Krista Fischer.
19.03.19 Sven Erik Ojavee (University of Lausanne, Switzerland) Elukestuse modelleerimine Bayesi Weibulli mudeliga komplekshaiguste korral/ Modelling survival and age-to-diagnosis with Bayesian Weibull model for complex diseases
05.03.19 Merli Mändul Models for the parental survival using offspring genotypes
26.02.19 Mare Vähi, Helle Visk Registrite kasutamisest jooksva statistika tegemisel ja registripõhises loenduses
19.02.19 Jüri Lember Evolutsioonimudeli(te)st II
12.02.19 Fabio Zucca (Milaano Polütehnikum) We generalize the evolution model introduced by Guiol, Machado and Schinazi (2010). In our model at odd times a random number X of species is created. Each species is endowed with a random fitness with arbitrary distribution on [0,1]. At even times a random number Y of species is removed, killing the species with lower fitness. We show that there is a critical fitness f_c below which the number of species hits zero i.o.~and above of which this number goes to infinity. We prove uniform convergence for the distribution of surviving species and describe the phenomena which could not be observed in previous works with uniformly distributed fitness.
18.12.2018 Kristi Kuljus Talk about the maximum spacing estimation method. The maximum spacing method is a parameter estimation method for continuous distributions
11.12.2018 Tõnu Kollo Mitmemõõtmeline ebasümmeetriline t-jaotus
04.12.2018 Imbi Traat

Visit to the University of Utrecht. Joint interests. Surprises in estimation under nonresponse

27.11.2018 Marika Kaakinen (Research Associate in Statistical Multiomics Modelling,  Imperial College, London, UK) Multi-phenotype, multi-omics and machine learning
20.11.18 Krista Fischer Mendelian Randomization
13.11.18 Mart Kals PhD thesis: Computational and statistical methods for analysing 2nd generation DNA sequencing data, with applications in the Estonian Biobank cohort
06.11.18 Kristi Läll

Genetic risk score for breast cancer

30.10.18 Raul Kangro EP projektist ja SNK tegemistest
23.10.18 Ene-Margit Tiit Indeksipõhine rahvastikustatistika käsitlus


Mohammad Jamsher Ali

Wider Bootstrap Confidence Interval. The estimation of the unknown parameter is the typical problem in applied statistics. It is obvious to raise the questions that 

* what estimator should be used and

* how accurate is the chosen estimator as an estimation of our parameter

The bootstrap is the general methodology to examine the accuracy of our estimator. It is a computer-based method first introduced by B. Efron in 1979. 

According to Efron, using skewness of the estimator it is possible to construct bootstrap confidence interval which is wider than standard confidence interval.


Maarika Traat

Pulsatility of small cortical veins using phase contrast magnetic resonance imaging at 7 Tesla.

Venous pulsatility measured from large cerebral veins has been shown to change in pathologies such as normal pressure hydrocephalus, Alzheimer’s disease, vascular dementia, mild cognitive impairment and multiple sclerosis. Venous pulsatility also increases with age. Recent improvements in the resolution of imaging techniques have made it feasible to explore smaller vessels. In my talk I give an overview of my Master’s project at the brain imaging centre of Cardiff University where, for the first time, we demonstrated the pulsatility of small cortical veins using phase-contrast magnetic resonance imaging at 7 tesla. Another very important finding was the statistically significant lag detected between the arrival of the cardiac pulse wave into the small cortical arteries and the small cortical veins. Measurements from smaller veins enhance the model of cerebral haemodynamics and add valuable insight into the medical conditions where intracranial pressure or pulsatility has been compromised, as well as into normal aging.



Konverentsidest, millel osaletud


Annika Krutto

valmiva doktoritöö "Empirical Cumulant Function Based Estimation in Stable Laws" tutvustamine. Juhendaja Tõnu Kollo.


Joonas Sova



Kaur Lumiste

valmiva doktoritöö Improving accuracy of survey estimators by using auxiliary information in data collection and estimation stages tutvustamine. Juhendaja Imbi Traat.


Paul Tammo

valmiva doktoritöö Closed maximal regular one-sided ideals in topological algebras tutvustamine. Juhendajad Mart Abel ja Mati Abel.


Annika Krutto

Stabiilse jaotuse parameetrite hindamine kumulantfunktsioonist. Stabiilsed jaotused moodustavad laia 4-parameetrilise tõenäosusjaotuste klassi. See eristati 1920ndatel kui sõltumatute sama jaotusega juhuslike suuruste normeeritud summa piirjaotus. Klass hõlmab nii kergete/raskete sabadega kui sümmeetrilised/asümmeetrilised juhud ja on leidnud ulatuslikku rakendust erinevates valdkondades. Stabiilse jaotuse parameetrite hindamise muudab keerukaks analüütilise jaotusfunktsiooni puudumine (välja arvatud erijuhud nagu  normaal-, Cauchy ja Levy jaotus). Viimastel aastakümnetel on  esitatud mitmeid lahendusi kuid neil kõigil on omad kitsendused. Käesoleva uurimus eesmärk on edasi arendada 1970ndatel välja pakutud üldistatud momentide meetodit.


Meelis Kull (LTAT)

Masinõppe algoritmide liigne enesekindlus ja kuidas seda vältida. Keerulised intelligentsed süsteemid nagu isesõitvad autod ja meditsiinilised ekspertsüsteemid rakendavad masinõppega ehitatud klassifikaatoreid. Tihti on tarvis, et need klassifikaatorid väljastaks koos ennustusega ka enesekindluse määra, sest see võimaldab süsteemil madala enesekindluse korral teha turvalisemaid valikuid. Seepärast on väga oluline, et klassifikaator ei oleks liiga enesekindel, sest see suurendaks väga kalliste vigade riski. Paraku on masinõppe meetoditega saadavad hinnangud siiski tihti liiga enesekindlad. Seminaris räägin millest see põhjustatud on ning kuidas kalibreerimise abil liigsest enesekindlusest lahti saada.