Search results for "tilasto"
showing 10 items of 142 documents
Calibrating Expert Assessments Using Hierarchical Gaussian Process Models
2020
Expert assessments are routinely used to inform management and other decision making. However, often these assessments contain considerable biases and uncertainties for which reason they should be calibrated if possible. Moreover, coherently combining multiple expert assessments into one estimate poses a long-standing problem in statistics since modeling expert knowledge is often difficult. Here, we present a hierarchical Bayesian model for expert calibration in a task of estimating a continuous univariate parameter. The model allows experts' biases to vary as a function of the true value of the parameter and according to the expert's background. We follow the fully Bayesian approach (the s…
Efficient estimation of generalized linear latent variable models.
2019
Generalized linear latent variable models (GLLVM) are popular tools for modeling multivariate, correlated responses. Such data are often encountered, for instance, in ecological studies, where presence-absences, counts, or biomass of interacting species are collected from a set of sites. Until very recently, the main challenge in fitting GLLVMs has been the lack of computationally efficient estimation methods. For likelihood based estimation, several closed form approximations for the marginal likelihood of GLLVMs have been proposed, but their efficient implementations have been lacking in the literature. To fill this gap, we show in this paper how to obtain computationally convenient estim…
How to reach optimal estimates of confidence intervals in microscopic counting of phytoplankton?
2021
Abstract Present practices in the microscopic counting of phytoplankton to estimate the reliability of results rely on the assumption of a random distribution of taxa in sample preparations. In contrast to that and in agreement with the literature, we show that aggregated distribution is common and can lead to over-optimistic confidence intervals, if estimated according to the shortcut procedure of Lund et al. based on the number of counted cells. We found a good linear correlation between the distribution independent confidence intervals for medians and those for parametric statistics so that 95% confidence intervals can be approximated by using a correction factor of 1.4. Instead, the rec…
Alkoholinkäytön aiheuttama liikennekuoleman riski Suomessa
2023
Tässä tutkimuksessa analysoidaan juopumuksen asteen vaikutusta kuolemaan johtavan liikenneonnettomuuden riskiin moderneilla tilastollisilla menetelmillä. Tutkimuksen pääaineistona ovat vuosina 2005–2014 Suomessa tapahtuneet kuolemaan johtaneet moottoriajoneuvo-onnettomuudet, joissa oli mukana henkilöauto aiheuttajana tai muuna osallisena. Pääaineistoa täydennetään ratsiatutkimuksista julkaistuilla tilastoilla. Keskeisiä käytettyjä mittareita ovat riskisuhde sekä vakioitu riskisuhde, josta on poistettu sekoittavien muuttujien vaikutusta. Tutkimuksen ensimmäisenä tavoitteena on selvittää, miten pieni veren alkoholipitoisuus alkaa lisätä liikenneonnettomuuden riskiä, ja toisena tavoitteena on …
Perusjoukon kokonaissumman estimointi Coxin regressiolla ja logistisella regressiolla : sovellus rakennusten aloitustilastoon
2012
A practical approach to improve the statistical performance of surface water monitoring networks
2019
The representativeness of aquatic ecosystem monitoring and the precision of the assessment results are of high importance when implementing the EU’s Water Framework Directive that aims to secure a good status of waterbodies in Europe. However, adapting monitoring designs to answer the objectives and allocating the sampling resources effectively are seldom practiced. Here, we present a practical solution how the sampling effort could be re-allocated without decreasing the precision and confidence of status class assignment. For demonstrating this, we used a large data set of 272 intensively monitored Finnish lake, coastal, and river waterbodies utilizing an existing framework for quantifying…
After the Millennium Development Goals. Remarks on the ethical assessment of global poverty reduction success
2018
The Millennium Development Goals were effective from 2000 to 2015. Statistics show that most of the goals were met, and particularly success in the goal of reducing extreme poverty (MDG1) gained wide recognition. Despite the strong ethical language related to poverty reduction, there has been little analysis of the ethical significance of the MDG achievements. Since statistical and ethical definitions and representations of poverty never completely overlap, conclusions concerning ethical progress are not directly available from the statistics. This article shows how this ethical significance can be analysed and what kinds of controversies and uncertainties relate to the issue. As part of th…
The value of perfect and imperfect information in lake monitoring and management.
2020
Highlights • Knowledge on the value of monitoring can assist decision-making in lake management. • We calculate value of perfect information theoretically. • We estimate value of imperfect information with Monte Carlo type of approach. • Generally, monitoring is profitable to invest in if VOI exceeds the cost. • Additional monitoring is profitable even if the lake is in good condition a priori. Uncertainty in the information obtained through monitoring complicates decision making about aquatic ecosystems management actions. We suggest the value of information (VOI) to assess the profitability of paying for additional monitoring information, when taking into account the costs and benefits of…
On the statistics of pairs of logarithms of integers
2022
We study the statistics of pairs of logarithms of positive integers at various scalings, either with trivial weights or with weights given by the Euler function, proving the existence of pair correlation functions. We prove that at the linear scaling, which is not the usual scaling by the inverse of the average gap, the pair correlations exhibit a level repulsion similar to radial distribution functions of fluids. We prove total loss of mass phenomena at superlinear scalings, and constant nonzero asymptotic behavior at sublinear scalings. The case of Euler weights has applications to the pair correlation of the lengths of common perpendicular geodesic arcs from the maximal Margulis cusp nei…
Requirement analysis for an artificial intelligence model for the diagnosis of the COVID-19 from chest X-ray data
2021
There are multiple papers published about different AI models for the COVID-19 diagnosis with promising results. Unfortunately according to the reviews many of the papers do not reach the level of sophistication needed for a clinically usable model. In this paper I go through multiple review papers, guidelines, and other relevant material in order to generate more comprehensive requirements for the future papers proposing a AI based diagnosis of the COVID-19 from chest X-ray data (CXR). Main findings are that a clinically usable AI needs to have an extremely good documentation, comprehensive statistical analysis of the possible biases and performance, and an explainability module.