Maximum likelihood estimation
Dotaz
Zobrazit nápovědu
Although the likelihood ratio is a well-known statistical technique, commercial off-the-shelf (COTS) software products for its calculation are not sufficiently validated to suit general requirements for the competence of testing and calibration laboratories (EN/ISO/IEC 17025:2005 norm) per se. The software in question can be considered critical as it directly weighs the forensic evidence allowing judges to decide on guilt or innocence or to identify person or kin (i.e.: in mass fatalities). For these reasons, accredited laboratories shall validate likelihood ratio software in accordance with the above norm. To validate software for calculating the likelihood ratio in parentage/kinship scenarios I assessed available vendors, chose two programs (Paternity Index and familias) for testing, and finally validated them using tests derived from elaboration of the available guidelines for the field of forensics, biomedicine, and software engineering. MS Excel calculation using known likelihood ratio formulas or peer-reviewed results of difficult paternity cases were used as a reference. Using seven testing cases, it was found that both programs satisfied the requirements for basic paternity cases. However, only a combination of two software programs fulfills the criteria needed for our purpose in the whole spectrum of functions under validation with the exceptions of providing algebraic formulas in cases of mutation and/or silent allele.
Functional diversity (FD) has the potential to address many ecological questions, from impacts of global change on biodiversity to ecological restoration. There are several methods estimating the different components of FD. However, most of these methods can only be computed at limited spatial scales and cannot account for intraspecific trait variability (ITV), despite its significant contribution to FD. Trait probability density (TPD) functions (which explicitly account for ITV) reflect the probabilistic nature of niches. By doing so, the TPD approach reconciles existing methods for estimating FD within a unifying framework, allowing FD to be partitioned seamlessly across multiple scales (from individuals to species, and from local to global scales), and accounting for ITV. We present methods to estimate TPD functions at different spatial scales and probabilistic implementations of several FD concepts, including the primary components of FD (functional richness, evenness, and divergence), functional redundancy, functional rarity, and solutions to decompose beta FD into nested and unique components. The TPD framework has the potential to unify and expand analyses of functional ecology across scales, capturing the probabilistic and multidimensional nature of FD. The R package TPD (https://CRAN.R-project.org/package=TPD) will allow users to achieve more comparative results across regions and case studies.
- MeSH
- biodiverzita * MeSH
- ekologie * MeSH
- fenotyp MeSH
- pravděpodobnostní funkce MeSH
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH
Verifying the speaker of a speech fragment can be crucial in attributing a crime to a suspect. The question can be addressed given disputed and reference speech material, adopting the recommended and scientifically accepted likelihood ratio framework for reporting evidential strength in court. In forensic practice, usually, auditory and acoustic analyses are performed to carry out such a verification task considering a diversity of features, such as language competence, pronunciation, or other linguistic features. Automated speaker comparison systems can also be used alongside those manual analyses. State-of-the-art automatic speaker comparison systems are based on deep neural networks that take acoustic features as input. Additional information, though, may be obtained from linguistic analysis. In this paper, we aim to answer if, when and how modern acoustic-based systems can be complemented by an authorship technique based on frequent words, within the likelihood ratio framework. We consider three different approaches to derive a combined likelihood ratio: using a support vector machine algorithm, fitting bivariate normal distributions, and passing the score of the acoustic system as additional input to the frequent-word analysis. We apply our method to the forensically relevant dataset FRIDA and the FISHER corpus, and we explore under which conditions fusion is valuable. We evaluate our results in terms of log likelihood ratio cost (Cllr) and equal error rate (EER). We show that fusion can be beneficial, especially in the case of intercepted phone calls with noise in the background.
- MeSH
- akustika řeči MeSH
- algoritmy MeSH
- lidé MeSH
- lingvistika MeSH
- pravděpodobnostní funkce MeSH
- řeč MeSH
- soudní vědy * metody MeSH
- support vector machine MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
This work presents the results of (137)Cs content long-term monitoring (1986-2007) in selected groups of foodstuffs. The data in the period of 1992-2007 were roughly log-normally distributed (beef, pork, milk). The more detailed statistical evaluation in this period of time was performed. Maximum likelihood method was used for estimation of the mean values. Several methods for decreasing the detection limits of (137)Cs (detector efficiency, sample preparation, counting containers) are listed and compared.
PURPOSE: To correlate the analysis of the optic nerve head (ONH) obtained with Disc Damage Likelihood Scale (DDLS), Heidelberg Retina Tomograph II (HRT II), and visual field (VF) parameters (Aulhorn scale, mean deviation [MD], and pattern standard deviation [PSD]) in patients with glaucoma. METHODS: Sixty-five eyes were evaluated using the DDLS classification. Fifty-seven eyes had stable primary open-angle glaucoma (POAG) and eight eyes were normal (control group). Classification of the ONH was performed using the HRT II, with Moorfields Regression Analysis (MRA) and Cup Shape Measure (CSM) indexes. The VF was examined by Full Threshold test (30-2) using the Humphrey Field Analyzer (HFA) perimeter and evaluated by three METHODS: Aulhorn's scale of 0 to 6, and two global indices, MD and PSD. RESULTS: The amount of ONH damage as estimated by the DDLS correlated well with topographic damage determined by HRT II (r=0.923 with MRA and r=0.807 with CSM) and with visual field loss as measured by the MD (r=-0.792) and PSD (r=0.718) of the HFA and Aulhorn field staging system (r=0.849). CONCLUSIONS: This study shows that evaluation of the ONH using the DDLS provides information that correlates well with VF and with data obtained from HRT II.
- MeSH
- diagnostické techniky oftalmologické MeSH
- discus nervi optici patologie MeSH
- dospělí MeSH
- financování organizované MeSH
- glaukom s otevřeným úhlem diagnóza MeSH
- lidé středního věku MeSH
- lidé MeSH
- nemoci zrakového nervu diagnóza MeSH
- nitrooční tlak MeSH
- poruchy zraku diagnóza MeSH
- pravděpodobnostní funkce MeSH
- regresní analýza MeSH
- senioři MeSH
- tomografie metody MeSH
- tonometrie oční MeSH
- zraková ostrost MeSH
- zraková pole MeSH
- Check Tag
- dospělí MeSH
- lidé středního věku MeSH
- lidé MeSH
- mužské pohlaví MeSH
- senioři MeSH
- ženské pohlaví MeSH
The ReAct (Recovery, Activity) project is an ENFSI (European Network of Forensic Science Institutes) supported initiative comprising a large consortium of laboratories. Here, the results from more than 23 laboratories are presented. The primary purpose was to design experiments simulating typical casework circumstances; collect data and to implement Bayesian networks to assess the value (i.e., likelihood ratio) of DNA results given activity level propositions. Two different experimental designs were used to simulate a robbery, where a screwdriver was used to force a door or window. Propositions and case information were chosen following laboratory feedback listing typical casework circumstances (included in the paper). In a direct transfer experiment, the defendant owned and used the screwdriver, but he did not force the door/window in question. An unknown person used the defendant's stolen screwdriver. In an indirect transfer experiment, the defendant neither owned, saw, nor used the screwdriver, nor did they force the door or window. For the second experiment, given the defence view, the defendant never held the screwdriver. We envisaged the situation where an object manipulated by the defendant (or the defendant himself/herself) would be touched by the unknown offender who would then force the window. It was found for the direct transfer experiment that unless a single contributor profile aligning with the known person's of interest profile was retrieved, the results did not allow to discriminate between propositions. On the other hand, for the indirect transfer experiment, both single and major contributor profiles that aligned with the person of interest (POI) supported the proposition that the person used the tool rather than an unknown person who had touched an object, when indeed the former was true. There was considerable variation in median recoveries of DNA between laboratories (between 200pg-5ng) for a given experiment if quantities are taken into account. These differences affect the likelihood ratios given activity level propositions. More than 2700 samples were analysed in the course of this study. Two different Bayesian Networks are made available via an open source application written in Shiny R: Shiny_React(). For comparison, all datasets were analysed using a qualitative method categorised into absent, single, major or other given contributors. The importance of standardising methods is emphasised, alongside the necessity of developing new approaches to assign the probability of laboratory-dependent DNA recovery. Freely accessible open databases play a crucial role in supporting these efforts.
- MeSH
- Bayesova věta * MeSH
- DNA fingerprinting * MeSH
- DNA * genetika MeSH
- laboratoře * MeSH
- lidé MeSH
- mikrosatelitní repetice MeSH
- pravděpodobnostní funkce MeSH
- soudní genetika metody MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
The refractory period concept is exploited in many neurophysiological studies as a part of developed methods. Computations connected to these methods are influenced by the estimated value of refractory period. In this article, parametric and nonparametric refractory period estimation methods (minimum, maximum likelihood estimate, minimum risk estimate, etc.) are compared for three neuronal models. The results acquired from simulated data illustrate both accuracy and bias of different techniques. Experimental data are also investigated to show the limitations of the methods discussed.
