arXiv subjects

Pauliina Ilmonen

Publications and source records attributed to Pauliina Ilmonen.

At least 19 recordsLinked to original sources

arXiv2026

Cellular Predictions on the Move: What about Data?

Mobile cellular load forecasting is native to network resource optimization and delivery of services with reliability, latency and quality guarantees. The mainstream of machine learning research in the area is focused primarily on developing powerful learning structures for improved prediction accuracy. The data used for forecasting traditionally belong to the cellular domain and at most contain exogenous information about the surroundings of the base stations. We approach the prediction task from the perspective of data as a vital component of any data learning process. We hypothesize that substantial improvements could be achieved when the data inform on the processes that create the cellular load. Specifically, we propose to characterize the population dynamics -- the potential number of cellular traffic sources and their mobility -- in addition to employing historical time series of mobile data traffic. We validate our hypothesis for the rarely examined highway scenario. Comprehensive experiments show forecasting improvements on the order of $60\%$ due to the use of these data alone.

cs.LG↗

arXiv2026

Parameter estimation in generalized fractional neuronal models

We investigate a generalized stochastic fractional neuronal model combining fractional dynamics with correlated stochastic inputs. The proposed framework is described by a fractional differential equation driven by a latent stochastic process with stationary increments and mean-reverting structure. This formulation allows the inclusion of both short-range and long-range dependence structures and naturally produces non-exponential relaxation phenomena. The main goal is the development of a feasible parameter estimation procedure based on discrete observations of the neuronal state process. We propose a two-step methodology. First, the parameters governing the fractional dynamics are estimated by exploiting the asymptotic behavior of Mittag-Leffler functions near the origin. Subsequently, the latent stochastic input is reconstructed through fractional differentiation techniques, allowing the estimation of the parameters governing the hidden noise dynamics. We derive quantitative error bounds for the estimators and analyze the reconstruction error of the latent process under suitable regularity assumptions on the driving noise. In particular, the interplay between the order of the fractional derivative and the Hölder regularity of the noise process naturally emerges in the stability analysis of the reconstruction procedure. Finally, simulation studies illustrate the applicability of the proposed methodology and highlight the influence of memory effects and noise regularity on the quality of statistical inference. The results support the relevance of fractional stochastic analysis for the modeling and inference of neuronal systems with memory and correlated inputs.

math.ST↗

arXiv2026

Characterization of continuous stationary fields as generalized Ornstein-Uhlenbeck fields via multi-parameter Langevin equation and multiple Riemann-Stieltjes integration

In this article, we characterize continuous stationary fields via generalized Langevin dynamics. This gives natural connections between stationary fields, stationary increment fields, self-similar fields, and generalized Langevin dynamics. Our contribution extends some recently proved similar results for stochastic processes to the case of continuous random fields. As a by-product, we introduce some new results on multiple Riemann-Stieltjes integrals.

math.PR↗

arXiv2025

Non-parametric estimation of non-linear diffusion coefficient in parabolic SPDEs

In this article, we introduce a novel non-parametric predictor, based on conditional expectation, for the unknown diffusion coefficient function $σ$ in the stochastic partial differential equation $Lu = σ(u)\dot{W}$, where $L$ is a parabolic second order differential operator and $\dot{W}$ is a suitable Gaussian noise. We prove consistency and derive an upper bound for the error in the $L^p$ norm, in terms of discretization and smoothening parameters $h$ and $\varepsilon$. We illustrate the applicability of the approach and the role of the parameters with several interesting numerical examples.

math.ST↗

arXiv2025

Data driven modeling of multiple interest rates with generalized Vasicek-type models

The Vasicek model is a commonly used interest rate model, and there exist many extensions and generalizations of it. However, most generalizations of the model are either univariate or assume the noise process to be Gaussian, or both. In this article, we study a generalized multivariate Vasicek model that allows simultaneous modeling of multiple interest rates while making minimal assumptions. In the model, we only assume that the noise process has stationary increments with a suitably decaying autocovariance structure. We provide estimators for the unknown parameters and prove their consistencies. We also derive limiting distributions for each estimator and provide theoretical examples. Furthermore, the model is tested empirically with both simulated data and real data.

econ.EM↗

arXiv2025

Extrapolated Markov Chain Oversampling Method for Imbalanced Text Classification

Text classification is the task of automatically assigning text documents correct labels from a predefined set of categories. In real-life (text) classification tasks, observations and misclassification costs are often unevenly distributed between the classes - known as the problem of imbalanced data. Synthetic oversampling is a popular approach to imbalanced classification. The idea is to generate synthetic observations in the minority class to balance the classes in the training set. Many general-purpose oversampling methods can be applied to text data; however, imbalanced text data poses a number of distinctive difficulties that stem from the unique nature of text compared to other domains. One such factor is that when the sample size of text increases, the sample vocabulary (i.e., feature space) is likely to grow as well. We introduce a novel Markov chain based text oversampling method. The transition probabilities are estimated from the minority class but also partly from the majority class, thus allowing the minority feature space to expand in oversampling. We evaluate our approach against prominent oversampling methods and show that our approach is able to produce highly competitive results against the other methods in several real data examples, especially when the imbalance is severe.

cs.LG↗

arXiv2025

Data Matters: The Case of Predicting Mobile Cellular Traffic

Accurate predictions of base stations' traffic load are essential to mobile cellular operators and their users as they support the efficient use of network resources and allow delivery of services that sustain smart cities and roads. Traditionally, cellular network time-series have been considered for this prediction task. More recently, exogenous factors such as points of interest and other environmental knowledge have been explored too. In contrast to incorporating external factors, we propose to learn the processes underlying cellular load generation by employing population dynamics data. In this study, we focus on smart roads and use road traffic measures to improve prediction accuracy. Comprehensive experiments demonstrate that by employing road flow and speed, in addition to cellular network metrics, base station load prediction errors can be substantially reduced, by as much as $56.5\%.$ The code, visualizations and extensive results are available on https://github.com/nvassileva/DataMatters.

cs.NI↗

arXiv2025

Moment Estimator-Based Extreme Quantile Estimation with Erroneous Observations: Application to Elliptical Extreme Quantile Region Estimation

In many application areas of extreme value theory, the variables of interest are not directly observable but instead contain errors. In this article, we quantify the effect of these errors in moment-based extreme value index estimation, and in corresponding extreme quantile estimation. We consider all, short-, light-, and heavy-tailed distributions. In particular, we derive conditions under which the error is asymptotically negligible. As an application, we consider affine equivariant extreme quantile region estimation under multivariate elliptical distributions.

math.ST↗

arXiv2025

Cut-off phenomenon and asymptotic mixing for multivariate general linear processes

The small noise cut-off phenomenon in continuous time and space has been studied in the recent literature for the linear and non-linear stable Langevin dynamics with additive Lévy drivers - understood as abrupt thermalization of the system along a particular time scale to its dynamical equilibrium - both for the total variation distance and the Wasserstein distance. The main result of this article establishes sufficient conditions for the window and profile cut-off phenomenon, which are flexible enough to cover the renormalized (non-Markovian) Ornstein--Uhlenbeck process driven by fractional Brownian motion and a large class of Gaussian and non-Gaussian, homogeneous and non-homogeneous drivers with (possible) finite second moments. The sufficient conditions are stated both for the total variation distance and the Wasserstein distance. Important examples are the multidimensional fractional Ornstein--Uhlenbeck process, the empirical sampling process of a fractional Ornstein--Uhlenbeck process, an Ornstein--Uhlenbeck processes driven by an Ornstein--Uhlenbeck process and the inhomogeneous Ornstein--Uhlenbeck process arising in simulated annealing.

math.PR↗

arXiv2024

On optimal prediction of missing functional data with memory

This paper considers the problem of reconstructing missing parts of functions based on their observed segments. It provides, for Gaussian processes and arbitrary bijective transformations thereof, theoretical expressions for the $L^2$-optimal reconstruction of the missing parts. These functions are obtained as solutions of explicit integral equations. In the discrete case, approximations of the solutions provide consistent expressions of all missing values of the processes. Rates of convergence of these approximations, under extra assumptions on the transformation function, are provided. In the case of Gaussian processes with a parametric covariance structure, the estimation can be conducted separately for each function, and yields nonlinear solutions in presence of memory. Simulated examples show that the proposed reconstruction indeed fares better than the conventional interpolation methods in various situations.

math.ST↗

arXiv2024

On the Impact of Approximation Errors on Extreme Quantile Estimation with Applications to Functional Data Analysis

We study the effect of approximation errors in assessing the extreme behavior of heavy-tailed random objects. We give conditions for the approximation error such that the standard asymptotic results hold for the classical Hill estimator and the corresponding extreme quantile estimator. As an application, we consider the effect of discretization errors in the computation of the $L^p$-norms related to functional data. We approximate the norms both with Riemann sums and with Monte Carlo integration. We quantify connections between the number of observed functions, the number of discretization points, and the regularity of the underlying functions. In addition, we derive a new concentration inequality for order statistics. This, to the best of our knowledge, is the first Chernoff-type concentration inequality for order statistics presented in the literature that provides an explicit rate at which the ratio between order statistics and tail quantile function converges to one. In our application, the bound is used to provide concentration inequalities measuring the distance between the Hill estimator based on approximated norms and the one based on the true ones.

math.ST↗

arXiv2024

Zombie Epidemic -- on Modeling the Effect of Interventions

The recent COVID-19 pandemic has highlighted the need of studying extreme, life-threatening phenomena in advance. In this article, a zombie epidemic in Uusimaa region in Finland is modeled. A stochastic agent based simulation model is proposed and extensive simulations are conducted for this purpose. The model utilizes knowledge on defensive human behavior during crises. Studying the effects of a hypothetical zombie attack resembles examining the spread of deadly diseases and of rumors. A zombie attack is simulated in the most densely populated region in Finland. The region's exact population densities over its rasterized geographical map are utilized. Furthermore, the simulations are used to study the effect of implementing a (strict or partial) quarantine area in the epicenter. Computationally efficient Scala codes and video animations of the simulated epidemics are provided. The main findings emphasize the importance of implementing very strict measures, without delay, to stop the outbreak.

stat.AP↗

arXiv2023

On Lamperti transformation and characterisations of discrete random fields

In this article we characterise discrete time stationary fields by difference equations involving stationary increment fields and self-similar fields. This gives connections between stationary fields, stationary increment fields and, through Lamperti transformation, self-similar fields. Our contribution is a natural generalisation of recently proved results covering the case of stationary processes.

math.PR↗

arXiv2022

On sharp rate of convergence for discretisation of integrals driven by fractional Brownian motions and related processes with discontinuous integrands

We consider equidistant approximations of stochastic integrals driven by Hölder continuous Gaussian processes of order $H>\frac12$ with discontinuous integrands involving bounded variation functions. We give exact rate of convergence in the $L^1$-distance and provide examples with different drivers. It turns out that the exact rate of convergence is proportional to $n^{1-2H}$ that is twice better compared to the best known results in the case of discontinuous integrands, and corresponds to the known rate in the case of smooth integrands. The novelty of our approach is that, instead of using multiplicative estimates for the integrals involved, we apply change of variables formula together with some facts on convex functions allowing us to compute expectations explicitly.

math.PR↗

arXiv2022

Flexible transition probability model for assessing cost-effectiveness of breast cancer screening

Breast cancer is the most common cancer among Western women. Fortunately, organized screening has reduced breast cancer mortality and, consequently, the European Union has recommended screening with mammography for 50-69-year-old women. This recommendation is followed well in Europe. Widening the screening target age further is supported by conditional recommendations for 45-49- and 70-74-year-old women. However, before extending screening to new age groups, it's essential to carefully consider the benefits and costs locally as circumstances vary between different regions and/or countries. We propose a new approach to assess cost-effectiveness of breast cancer screening for a long-ongoing program with incomplete historical screening data. The new model is called flexible stage distribution model. It is based on estimating the stage distributions of breast cancer cases under different screening strategies. In the model, an ongoing screening strategy may be used as a baseline and other screening strategies may be incorporated by changes in the incidence rates. The model is flexible, as it enables to apply different approaches for estimating the altered stage distributions. Thus, if randomized data is available, one may rely on that. On the other hand, if randomized data is not available, altered stage distributions may be estimated by extrapolating the stage distributions of the youngest and oldest screened/non-screened age groups. We apply the proposed flexible stage distribution model for assessing incremental cost of extending the current biennial breast cancer screening to younger and older target ages in Finland.

stat.AP↗

arXiv2021

Integrated shape-sensitive functional metrics

This paper develops a new integrated ball (pseudo)metric which provides an intermediary between a chosen starting (pseudo)metric d and the L_p distance in general function spaces. Selecting d as the Hausdorff or Fréchet distances, we introduce integrated shape-sensitive versions of these supremum-based metrics. The new metrics allow for finer analyses in functional settings, not attainable applying the non-integrated versions directly. Moreover, convergent discrete approximations make computations feasible in practice.

math.MG↗

arXiv2020

Vector-valued Generalised Ornstein-Uhlenbeck Processes

Generalisations of the Ornstein-Uhlenbeck process defined through Langevin equation $dU_t = - ΘU_t dt + dG_t,$ such as fractional Ornstein-Uhlenbeck processes, have recently received a lot of attention in the literature. In particular, estimation of the unknown parameter $Θ$ is widely studied under Gaussian stationary increment noise $G$. Langevin equation is well-known for its connections to physics. In addition to that, motivation for studying Langevin equation with a general noise $G$ stems from the fact that the equation characterises all univariate stationary processes. Most of the literature on the topic focuses on the one-dimensional case with Gaussian noise $G$. In this article, we consider estimation of the unknown model parameter in the multidimensional version of the Langevin equation, where the parameter $Θ$ is a matrix and $G$ is a general, not necessarily Gaussian, vector-valued process with stationary increments. Based on algebraic Riccati equations, we construct an estimator for the matrix $Θ$. Moreover, we prove the consistency of the estimator and derive its limiting distribution under natural assumptions. In addition, to motivate our work, we prove that the Langevin equation characterises all stationary processes in a multidimensional setting as well.

math.ST↗

arXiv2020

Continuous time Gaussian process dynamical models in gene regulatory network inference

One of the focus areas of modern scientific research is to reveal mysteries related to genes and their interactions. The dynamic interactions between genes can be encoded into a gene regulatory network (GRN), which can be used to gain understanding on the genetic mechanisms behind observable phenotypes. GRN inference from time series data has recently been a focus area of systems biology. Due to low sampling frequency of the data, this is a notoriously difficult problem. We tackle the challenge by introducing the so-called continuous-time Gaussian process dynamical model, based on Gaussian process framework that has gained popularity in nonlinear regression problems arising in machine learning. The model dynamics are governed by a stochastic differential equation, where the dynamics function is modelled as a Gaussian process. We prove the existence and uniqueness of solutions of the stochastic differential equation. We derive the probability distribution for the Euler discretised trajectories and establish the convergence of the discretisation. We develop a GRN inference method called BINGO, based on the developed framework. BINGO is based on MCMC sampling of trajectories of the GPDM and estimating the hyperparameters of the covariance function of the Gaussian process. Using benchmark data examples, we show that BINGO is superior in dealing with poor time resolution and it is computationally feasible.

math.OC↗