1,721,017 research outputs found
Variational Inference for GARCH-family Models
Full text linkThe Bayesian estimation of GARCH-family models has been typically addressed through Monte Carlo sampling. Variational Inference is gaining popularity and attention as a robust approach for Bayesian inference in complex machine learning models; however, its adoption in econometrics and finance is limited. This paper discusses the extent to which Variational Inference constitutes a reliable and feasible alternative to Monte Carlo sampling for Bayesian inference in GARCH-like models. Through a large-scale experiment involving the constituents of the S& P 500 index, several Variational Inference optimizers, a variety of volatility models, and a case study, we show that Variational Inference is an attractive, remarkably well-calibrated, and competitive method for Bayesian learning
La terminografia orientata alla traduzione tra pragmatismo e armonizzazione
No full textThe articles focuses on one of the fundamental choices to be made prior to the creation of a translation-oriented terminological data bank, namely, its position between absolute prescription and absolute description in order to find the right balance between pragmatics and harmonisation. Both the traditional outlook on terminology, which emphasises the importance of standardisation, and some recent approaches , that lay stress on single texts, are unsuitable. The most promising solution seems to be an intermediate one, based on the analysis of a significant corpus of texts in order to determine the privileged characteristics and other prototypical factors: on the one hand, the importance for the specialised translator to obtain the pragmatic equivalent even on a terminological level, on the other, the contribution the translator can make towards harmonisation of terminology. The latter does not imply a prescriptive approach, but entails paying greater attention to the self-regulating attempts (either conscious or unconscious) of specialists
Vaghezza e connotazioni: elementi di disturbo nelle definizioni terminologiche?
Full text linkLe nuove correnti della terminologia, che hanno tratto beneficio da importanti apporti dello studio delle lingue speciali ma anche di altre discipline quali la linguistica cognitiva, la sociologia e la
filosofia della scienza, si sono staccate da alcuni dei principi cardine della terminologia tradizionale di stampo prescrittivo ed hanno “riabilitato” concetti a lungo banditi dalla letteratura del settore. Il presente articolo intende focalizzarsi su due di questi concetti, vaghezza e connotazioni, per riesaminarne il ruolo
nelle definizioni da una prospettiva sia terminologica che terminografica. Sulla scorta di esempi tratti da diverse discipline si illustrerà come nelle definizioni vaghezza e connotazioni non siano infrequenti e abbiano sovente una motivazione ben precisa. Sotto il profilo terminografico si evidenzierà l’importanza di
conciliare analisi sincronica e diacronica e di integrare la definizione con informazioni fornite tramite contesti, note o il campo dedicato all’equivalenza interlinguistica, in modo da agevolare l’utente della raccolta terminografica nella comprensione di concetti spesso non banali
Strategic ejaculate adjustments and mismatches: are males paying sperm senescence costs?
No full textIn many species, males show anticipatory plasticity for sperm production, which they adjust to match perceived mating opportunities. While the strategic adjustment of sperm production is likely to be beneficial, it may be also associated with costs, including those arising from the expression of a phenotype that is poorly matched to the conditions that males will subsequently experience. Mismatch costs are exacerbated by trade-offs between investment in the ejaculate and investment in other traits and by trade-offs among individual ejaculate traits. Trade-offs, in fact, may determine a decrease in male competitiveness, due to impaired ability to obtain copulations or to reduced ejaculate quality. We explored mismatch costs using male guppies, which are known to increase sperm production, but reduce their investment in sexual behaviour, when maintained in the presence of females. Increasing ejaculate size in the absence of females could impose costs that would be paid when an opportunity to mate eventually arises. One of such costs may involve male post-copulatory competitiveness and may be associated with increased sperm senescence. To explore mismatch costs, firstly we induced two groups of males to differentiate their sperm production by exposing them or not to female stimuli. Then, we isolated them to prevent matings and have their sperm ageing. Finally, we compared ejaculate quality between the groups. Contrary to expectations, we found that female-stimulated males did not suffer from increased sperm senescence. These costs are probably minimized by the high level of plasticity associated with this trait, resulting in sperm production being quickly re-adjusted to a new environment. Other types of mismatch costs may be more relevant, for example, those related to trade-offs with sexual behaviour
Discussing otherness through the linguistic representation of gender: from feminism to the rejection of the gender binary
No full textThe concept of ‘Self’ as opposed to ‘Other’ can describe individual and group identity in a given socio-cultural context, where Self and Other can be applied to a range of categories, from ethnicity to social class. Dominant groups generally use their power to control categorisation systems, thus allowing some of the resulting ‘classes’ (e.g. males, whites, middle-class people, and heterosexuals) to be attributed a ‘standard’ status, while discriminating and marginalising groups with different characteristics. Non-standard identities are thus perceived as other. This work focuses on gender, considered as a key category which contributes to define individual and socio-cultural identity, with a focus on Western English- and Italian-speaking contexts. As such, gender has been employed and linguistically represented as a means of categorisation where groups and identities are evaluated and treated according to dominant man-centred and heteronormative views. Groups generally perceived as ‘normal’ are thus separated from groups labelled as other, which fall outside the standards. Gendered Self and Other have assumed different shapes and configurations in time, and this has been reflected (and reflected upon) through language. In this work, the relation between gender and language is analysed with reference to previous studies, particularly by taking into account instances of gender-based oppression and discrimination, represented and carried out through language. In such context, the perceptions of Self and Other take place at different levels and across different communities. However, the concept of gender itself has been questioned, by introducing the possibility of rejecting or re-elaborating its traditionally binary “male or female” option. This produces new distinctions between norm-conforming self and non-binary, norm-defying other. This potential subversion of binary gender notions can have linguistic implications. Of these, two have been analysed in particular: the first, more general, is the attempt at promoting a more inclusive, gender neutral way to refer to people; the second, more specific, is the way in which English language media recently represented the refusal, by a part of the LGBTQ community, as well as by people who reject any such label, to be categorised and referred to as either male or female
Introduzione
No full textL'opera fornisce un'introduzione ai metodi di descrizione e gestione della terminologia, una disciplina nuova e di sicuro interesse data l'importanza crescente delle lingue speciali e la conseguente necessità di banche dati e dizionari specializzati. Il testo nasce dal contributo di vari esperti italiani e stranieri che da anni si occupano in ambito accademico e professionale dei problemi connessi alla terminologia e terminografia. Rappresenta la prima pubblicazione sistematica sulla materia in lingua italiana. Data la struttura articolata in agili capitoli su aspetti specifici, il volume costituisce non solo un valido testo didattico per i corsi di traduzione e quellii di lingua per gli studenti di discipline scientifiche, ma anche un'utile opera di riferimento per gli operatori che gestiscono banche dati terminografiche
The rhythm of the Other. Yves Bonnefoy and Philippe Jaccottet between poetry and translation
No full textUncertainty Estimation in Deep Bayesian Survival Models
No full textBayesian methods can express uncertainty about their predictions, but have seen little adaptation in survival analysis using neural networks. Proper uncertainty estimation is important in high-risk domains, such as the healthcare or medical field, if machine learning methods are to be adopted for decision-making purposes, however, uncertainty estimation is a known shortcoming of neural networks. In this paper, we introduce the use of Bayesian inference techniques for survival analysis in neural networks that rely on the Cox proportional hazard assumption, for which we discuss a new flexible and effective architecture. We implement three architectures: a fully-deterministic neural network that acts as a baseline, a Bayesian model using variational inference, and one using Monte-Carlo Dropout. Our comprehensive experiments show that on the WHAS500 dataset, Bayesian techniques improve predictive performance over the state-of-the-art neural networks and on the larger SEER and SUPPORT datasets provide comparable performance. In all experiments, training with Monte Carlo Dropout is significantly faster than training with variational inference. Our Bayesian models additionally provide quantification of both aleatoric and epistemic uncertainty, which we exhibit by plotting 95% confidence intervals around the survival function and showing a probability density function of the survival time. Our work motivates further work in leveraging uncertainty for survival analysis using neural networks
Benchmark dataset for mid-price forecasting of limit order book data with machine learning methods
No full textManaging the prediction of metrics in high-frequency financial markets is a challenging task. An efficient way is by monitoring the dynamics of a limit order book to identify the information edge. This paper describes the first publicly available benchmark dataset of high-frequency limit order markets for mid-price prediction. We extracted normalized data representations of time series data for five stocks from the Nasdaq Nordic stock market for a time period of 10 consecutive days, leading to a dataset of ∼4,000,000 time series samples in total. A day-based anchored cross-validation experimental protocol is also provided that can be used as a benchmark for comparing the performance of state-of-the-art methodologies. Performance of baseline approaches are also provided to facilitate experimental comparisons. We expect that such a large-scale dataset can serve as a testbed for devising novel solutions of expert systems for high-frequency limit order book data analysis
- …
