Ratio-based Approach Research Articles

Optimisation of nominal-the-best type quality characteristic in the absence of scaling factor: an empirical method

For a nominal-the-best type of response variable, Taguchi's two-step optimisation process uses signal-to-noise (SN) ratio for minimising the variability in the first stage and then adjusts the mean to the target by changing the levels of some scaling factor that affect only the mean and not the variability. In reality, the input variables often affect both the variability and mean value. In such situations, Taguchi's SN ratio-based optimisation approach is not truly applicable. There prevail some response surface methodology-based approaches which can tackle the above situations. However, fitting well-explained response surfaces for mean and variance is not quite an easy task for many quality practitioners, particularly those who do not have a strong background in mathematics/statistics. In this paper, a simple empirical method is proposed, which requires neither a scaling factor nor response surfaces for mean and variance. The performance of the proposed method is compared with other methods using three case studies. The results show that the proposed method results in optimal or very close to optimal solution.

Searching for first-degree familial relationships in California's offender DNA database: Validation of a likelihood ratio-based approach

A validation study was performed to measure the effectiveness of using a likelihood ratio-based approach to search for possible first-degree familial relationships (full-sibling and parent–child) by comparing an evidence autosomal short tandem repeat (STR) profile to California's ∼1,000,000-profile State DNA Index System (SDIS) database. Test searches used autosomal STR and Y-STR profiles generated for 100 artificial test families. When the test sample and the first-degree relative in the database were characterized at the 15 Identifiler ® (Applied Biosystems ®, Foster City, CA) STR loci, the search procedure included 96% of the fathers and 72% of the full-siblings. When the relative profile was limited to the 13 Combined DNA Index System (CODIS) core loci, the search procedure included 93% of the fathers and 61% of the full-siblings. These results, combined with those of functional tests using three real families, support the effectiveness of this tool. Based upon these results, the validated approach was implemented as a key, pragmatic and demonstrably practical component of the California Department of Justice's Familial Search Program. An investigative lead created through this process recently led to an arrest in the Los Angeles Grim Sleeper serial murders.

Combining spatial data in landslide reactivation susceptibility mapping: A likelihood ratio-based approach in W Belgium

A key issue in landslide susceptibility mapping concerns the relevance of the spatial data combination used in the prediction. Various combinations of high-resolution predictor variables and possibilities of selecting them from a larger dataset are analysed. The scarp reactivation of several landslides in a hilly region of W Belgium is investigated at the pixel scale. The susceptibility modelling uses the reactivated scarp segments as the dependent variable and 13 factors at a 2 m-resolution related to topography, hydrology, land use and lithology as potential independent variables. The modelling uses a likelihood ratio approach based on the comparison, for each independent variable, between two empirical distribution functions (EDFs), respectively for the reactivated and non-reactivated areas. It uses these EDFs as favourability values to build membership values and combine them with a fuzzy Gamma operator. Five different data combinations are tested and compared by analysing the prediction-rate curves obtained by cross-validation. The geomorphological value of the resulting susceptibility maps is also discussed. This research shows relevant results for predicting the susceptibility to scarp reactivation. We highlight the need for testing several data combinations and underline that combining quantitative criteria with expert opinion is an asset for reliable predictions.

System Uncertainty and Statistical Detection for Jump-diffusion Models

Motivated by the common-seen model uncertainty of real-world systems, we propose a likelihood ratio-based approach to statistical detection for a rich class of partially observed systems. Here, the system state is modeled by some jump-diffusion process while the observation is of additive white noise. Our approach can be implemented recursively based on some Markov chain approximation method to compare the competing stochastic models by fitting the observed historical data. Our method is superior to the traditional hypothesis test in both theoretical and computational aspects. In particular, a wide range of different models can be nested and compared in a unified framework with the help of Bayes factor. An illustrating numerical example is also given to show the application of our method.

Exploring the Discriminatory Potential of F0 Distribution Parameters in Traditional Forensic Speaker Recognition

Despite its many prima facie attractive properties for Forensic Speaker Recognition, F0 is regarded as having limited forensic value due to its large within-speaker variability. However, its forensic use to date has been limited mostly to its long-term mean and standard deviation. This paper examines the discriminatory potential, within a Likelihood Ratio-based approach, of additional parametric features from the distribution of long-term F0: its skew, kurtosis, modal F0 and modal density. Motivated by the observation that the shape of the long-term F0 distribution shows less within-speaker occasion-to-occasion difference, we report a forensic discrimination experiment with non-contemporaneous speech samples from 201 male Japanese speakers. Using a multivariate Likelihood Ratio as discriminant distance with the six LTF0 distribution parameters, an equal error rate of 10.7% is obtained from 201 target and 80400 non-target trials. We also investigate how the EER degrades as a function of amount of voiced speech.

Statistical model uncertainty for state-observation models driven by diffusion process

This paper is motivated by the fact that model uncertainty is common-seen in statistics and applied probability. As a response, a likelihood ratio-based approach is presented to statistical detection within a class of state-observation models. Here, the system is driven by some diffusion process while the continuous-time observation process is of additive white noise. The proposed approach can be implemented recursively to compare the competing stochastic systems by fitting the observed historical data. It is superior to the traditional hypothesis test in both theoretical and numerical senses. Moreover, it can be extended to the more general martingale problem setup.

Absolute quantitation of intracellular metabolite concentrations by an isotope ratio-based approach

This protocol provides a method for quantitating the intracellular concentrations of endogenous metabolites in cultured cells. The cells are grown in stable isotope-labeled media to near-complete isotopic enrichment and then extracted in organic solvent containing unlabeled internal standards in known concentrations. The ratio of endogenous metabolite to internal standard in the extract is determined using mass spectrometry (MS). The product of this ratio and the unlabeled standard amount equals the amount of endogenous metabolite present in the cells. The cellular concentration of the metabolite can then be calculated on the basis of intracellular volume of the extracted cells. The protocol is exemplified using Escherichia coli and primary human fibroblasts fed uniformly with (13)C-labeled carbon sources, with detection of (13)C-assimilation by liquid chromatography-tandem MS. It enables absolute quantitation of several dozen metabolites over approximately 1 week of work.

An extended likelihood ratio framework for interpreting evidence

This paper reviews some current methods, the likelihood ratio-based approach and the full Bayesian approach for the interpretation of evidence and discusses previously identified shortcomings in them. It suggests an approach based on a compromise--based on an extended likelihood ratio--that may combine the merits of logic without overstepping acceptable bounds for the forensic scientist in the presentation of evidence. The approach is exposed formally and takes advantage of inferential networks called Bayesian networks.

Technical forensic speaker recognition: Evaluation, types and testing of evidence

Important aspects of Technical Forensic Speaker Recognition, particularly those associated with evidence, are exemplified and critically discussed, and comparisons drawn with generic Speaker Recognition. The centrality of the Likelihood Ratio of Bayes’ theorem in correctly evaluating strength of forensic speech evidence is emphasised, as well as the many problems involved in its accurate estimation. It is pointed out that many different types of evidence are of use, both experimentally and forensically, in discriminating same-speaker from different-speaker speech samples, and some examples are given from real forensic case-work to illustrate the Likelihood Ratio-based approach. The extent to which Technical Forensic Speaker Recognition meets the Daubert requirement of testability is also discussed.

Testing realistic forensic speaker identification in Japanese: a likelihood ratio-based approach using formants

Testing realistic forensic speaker identification in Japanese: a likelihood ratio-based approach using formants