Searcharxiv⌕ Search

arXiv subjects

Ioannis Kosmidis

Publications and source records attributed to Ioannis Kosmidis.

31 records · Page 2Linked to original sources
arXiv2019

Davidson-Luce model for multi-item choice with ties

This paper introduces a natural extension of the pair-comparison-with-ties model of Davidson (1970, J. Amer. Statist. Assoc.), to allow for ties when more than two items are compared. Properties of the new model are discussed. It is found that this "Davidson-Luce" model retains the many appealing features of Davidson's solution, while extending the scope of application substantially beyond the domain of pair-comparison data. The model introduced here already underpins the handling of tied rankings in the "PlackettLuce" R package.

stat.ME↗
arXiv2019

Mean and median bias reduction in generalized linear models

This paper presents an integrated framework for estimation and inference from generalized linear models using adjusted score equations that result in mean and median bias reduction. The framework unifies theoretical and methodological aspects of past research on mean bias reduction and accommodates, in a natural way, new advances on median bias reduction. General expressions for the adjusted score functions are derived in terms of quantities that are readily available in standard software for fitting generalized linear models. The resulting estimating equations are solved using a unifying quasi-Fisher scoring algorithm that is shown to be equivalent to iteratively re-weighted least squares with appropriately adjusted working variates. Formal links between the iterations for mean and median bias reduction are established. Core model invariance properties are used to develop a novel mixed adjustment strategy when the estimation of a dispersion parameter is necessary. It is also shown how median bias reduction in multinomial logistic regression can be done using the equivalent Poisson log-linear model. The estimates coming out from mean and median bias reduction are found to overcome practical issues related to infinite estimates that can occur with positive probability in generalized linear models with multinomial or discrete responses, and can result in valid inferences even in the presence of a high-dimensional nuisance parameter

stat.ME↗
arXiv2018

Modeling outcomes of soccer matches

We compare various extensions of the Bradley-Terry model and a hierarchical Poisson log-linear model in terms of their performance in predicting the outcome of soccer matches (win, draw, or loss). The parameters of the Bradley-Terry extensions are estimated by maximizing the log-likelihood, or an appropriately penalized version of it, while the posterior densities of the parameters of the hierarchical Poisson log-linear model are approximated using integrated nested Laplace approximations. The prediction performance of the various modeling approaches is assessed using a novel, context-specific framework for temporal validation that is found to deliver accurate estimates of the test error. The direct modeling of outcomes via the various Bradley-Terry extensions and the modeling of match scores using the hierarchical Poisson log-linear model demonstrate similar behavior in terms of predictive performance.

stat.AP↗
arXiv2018

Median bias reduction in random-effects meta-analysis and meta-regression

Random-effects models are frequently used to synthesise information from different studies in meta-analysis. While likelihood-based inference is attractive both in terms of limiting properties and of implementation, its application in random-effects meta-analysis may result in misleading conclusions, especially when the number of studies is small to moderate. The current paper shows how methodology that reduces the asymptotic bias of the maximum likelihood estimator of the variance component can also substantially improve inference about the mean effect size. The results are derived for the more general framework of random-effects meta-regression, which allows the mean effect size to vary with study-specific covariates.

stat.AP↗
arXiv2017

Improving the accuracy of likelihood-based inference in meta-analysis and meta-regression

Random-effects models are frequently used to synthesise information from different studies in meta-analysis. While likelihood-based inference is attractive both in terms of limiting properties and of implementation, its application in random-effects meta-analysis may result in misleading conclusions, especially when the number of studies is small to moderate. The current paper shows how methodology that reduces the asymptotic bias of the maximum likelihood estimator of the variance component can also substantially improve inference about the mean effect size. The results are derived for the more general framework of random-effects meta-regression, which allows the mean effect size to vary with study-specific covariates.

stat.ME↗
arXiv2015

Model-based clustering using copulas with applications

The majority of model-based clustering techniques is based on multivariate Normal models and their variants. In this paper copulas are used for the construction of flexible families of models for clustering applications. The use of copulas in model-based clustering offers two direct advantages over current methods: i) the appropriate choice of copulas provides the ability to obtain a range of exotic shapes for the clusters, and ii) the explicit choice of marginal distributions for the clusters allows the modelling of multivariate data of various modes (either discrete or continuous) in a natural way. This paper introduces and studies the framework of copula-based finite mixture models for clustering applications. Estimation in the general case can be performed using standard EM, and, depending on the mode of the data, more efficient procedures are provided that can fully exploit the copula structure. The closure properties of the mixture models under marginalization are discussed, and for continuous, real-valued data parametric rotations in the sample space are introduced, with a parallel discussion on parameter identifiability depending on the choice of copulas for the components. The exposition of the methodology is accompanied and motivated by the analysis of real and artificial data.

stat.ME↗
arXiv2015

Linking the performance of endurance runners to training and physiological effects via multi-resolution elastic net

A multiplicative effects model is introduced for the identification of the factors that are influential to the performance of highly-trained endurance runners. The model extends the established power-law relationship between performance times and distances by taking into account the effect of the physiological status of the runners, and training effects extracted from GPS records collected over the course of a year. In order to incorporate information on the runners' training into the model, the concept of the training distribution profile is introduced and its ability to capture the characteristics of the training session is discussed. The covariates that are relevant to runner performance as response are identified using a procedure termed multi-resolution elastic net. Multi-resolution elastic net allows the simultaneous identification of scalar covariates and of intervals on the domain of one or more functional covariates that are most influential for the response. The results identify a contiguous group of speed intervals between 5.3 to 5.7 m$\cdot$s$^{-1}$ as influential for the improvement of running performance and extend established relationships between physiological status and runner performance. Another outcome of multi-resolution elastic net is a predictive equation for performance based on the minimization of the mean squared prediction error on a test data set across resolutions.

stat.AP↗
arXiv2014

Liquidity commonality does not imply liquidity resilience commonality: A functional characterisation for ultra-high frequency cross-sectional LOB data

We present a large-scale study of commonality in liquidity and resilience across assets in an ultra high-frequency (millisecond-timestamped) Limit Order Book (LOB) dataset from a pan-European electronic equity trading facility. We first show that extant work in quantifying liquidity commonality through the degree of explanatory power of the dominant modes of variation of liquidity (extracted through Principal Component Analysis) fails to account for heavy tailed features in the data, thus producing potentially misleading results. We employ Independent Component Analysis, which both decorrelates the liquidity measures in the asset cross-section, but also reduces higher-order statistical dependencies. To measure commonality in liquidity resilience, we utilise a novel characterisation as the time required for return to a threshold liquidity level. This reflects a dimension of liquidity that is not captured by the majority of liquidity measures and has important ramifications for understanding supply and demand pressures for market makers in electronic exchanges, as well as regulators and HFTs. When the metric is mapped out across a range of thresholds, it produces the daily Liquidity Resilience Profile (LRP) for a given asset. This daily summary of liquidity resilience behaviour from the vast LOB dataset is then amenable to a functional data representation. This enables the comparison of liquidity resilience in the asset cross-section via functional linear sub-space decompositions and functional regression. The functional regression results presented here suggest that market factors for liquidity resilience (as extracted through functional principal components analysis) can explain between 10 and 40% of the variation in liquidity resilience at low liquidity thresholds, but are less explanatory at more extreme levels, where individual asset factors take effect.

q-fin.ST↗
arXiv2014

Upside and Downside Risk Exposures of Currency Carry Trades via Tail Dependence

Currency carry trade is the investment strategy that involves selling low interest rate currencies in order to purchase higher interest rate currencies, thus profiting from the interest rate differentials. This is a well known financial puzzle to explain, since assuming foreign exchange risk is uninhibited and the markets have rational risk-neutral investors, then one would not expect profits from such strategies. That is, according to uncovered interest rate parity (UIP), changes in the related exchange rates should offset the potential to profit from such interest rate differentials. However, it has been shown empirically, that investors can earn profits on average by borrowing in a country with a lower interest rate, exchanging for foreign currency, and investing in a foreign country with a higher interest rate, whilst allowing for any losses from exchanging back to their domestic currency at maturity. This paper explores the financial risk that trading strategies seeking to exploit a violation of the UIP condition are exposed to with respect to multivariate tail dependence present in both the funding and investment currency baskets. It will outline in what contexts these portfolio risk exposures will benefit accumulated portfolio returns and under what conditions such tail exposures will reduce portfolio returns.

q-fin.PM↗
arXiv2013

Bias in parametric estimation: reduction and useful side-effects

The bias of an estimator is defined as the difference of its expected value from the parameter to be estimated, where the expectation is with respect to the model. Loosely speaking, small bias reflects the desire that if an experiment is repeated indefinitely then the average of all the resultant estimates will be close to the parameter value that is estimated. The current paper is a review of the still-expanding repository of methods that have been developed to reduce bias in the estimation of parametric models. The review provides a unifying framework where all those methods are seen as attempts to approximate the solution of a simple estimating equation. Of particular focus is the maximum likelihood estimator, which despite being asymptotically unbiased under the usual regularity conditions, has finite-sample bias that can result in significant loss of performance of standard inferential procedures. An informal comparison of the methods is made revealing some useful practical side-effects in the estimation of popular models in practice including: i) shrinkage of the estimators in binomial and multinomial regression models that guarantees finiteness even in cases of data separation where the maximum likelihood estimator is infinite, and ii) inferential benefits for models that require the estimation of dispersion or precision parameters.

stat.ME↗
arXiv2013

Improved estimation in cumulative link models

For the estimation of cumulative link models for ordinal data, the bias-reducing adjusted score equations in \citet{firth:93} are obtained, whose solution ensures an estimator with smaller asymptotic bias than the maximum likelihood estimator. Their form suggests a parameter-dependent adjustment of the multinomial counts, which, in turn suggests the solution of the adjusted score equations through iterated maximum likelihood fits on adjusted counts, greatly facilitating implementation. Like the maximum likelihood estimator, the reduced-bias estimator is found to respect the invariance properties that make cumulative link models a good choice for the analysis of categorical data. Its additional finiteness and optimal frequentist properties, along with the adequate behaviour of related asymptotic inferential procedures make the reduced-bias estimator attractive as a default choice for practical applications. Furthermore, the proposed estimator enjoys certain shrinkage properties that are defensible from an experimental point of view relating to the nature of ordinal data.

stat.ME↗
arXiv2009

Simulating Events of Unknown Probabilities via Reverse Time Martingales

Assume that one aims to simulate an event of unknown probability $s\in (0,1)$ which is uniquely determined, however only its approximations can be obtained using a finite computational effort. Such settings are often encountered in statistical simulations. We consider two specific examples. First, the exact simulation of non-linear diffusions, second, the celebrated Bernoulli factory problem of generating an $f(p)-$coin given a sequence $X_1,X_2,...$ of independent tosses of a $p-$coin (with known $f$ and unknown $p$). We describe a general framework and provide algorithms where this kind of problems can be fitted and solved. The algorithms are straightforward to implement and thus allow for effective simulation of desired events of probability $s.$ In the case of diffusions, we obtain the algorithm of \cite{BeskosRobertsEA1} as a specific instance of the generic framework developed here. In the case of the Bernoulli factory, our work offers a statistical understanding of the Nacu-Peres algorithm for $f(p) = \min\{2p, 1-2\varepsilon\}$ (which is central to the general question) and allows for its immediate implementation that avoids algorithmic difficulties of the original version.

stat.CO↗