Short Data Samples Research Articles

CCDH: Complexity based Causal Discovery of Hidden common cause in time series

Causal discovery based on temporal observations poses a significant challenge, especially when dealing with causal relationships among dynamical systems exhibiting chaotic attractors. Existing solutions are data-intensive and unable to detect hidden common drivers (also known as common causes or confounders). To address these limitations, we propose a novel method that overcomes both issues. Our method relies on Takens’ embedding theorem and assesses the complexity of rank order patterns in the embedded series by measuring compressibility through non-sequential recursive pair substitution. Remarkably, this approach is effective even with short data samples and has the capability to detect both unidirectional and bidirectional causation, as well as hidden common causes. To validate its performance, we apply the method to synthetic datasets and human electrophysiological data obtained from an epileptic patient. The method successfully provides insights into the involvement of the left and right hippocampus before, during, and after an epileptic seizure. Consequently, our method may offer valuable additional information for decision-making medical panels in determining the optimal intervention locations. Due to its advantages and simplicity, this method holds promise for application in various scientific and practical domains with successful outcomes.

Packet Loss Compensation for VoIP through Bone‐Conducted Speech Using Modified Linear Prediction

In this paper, we compare air‐conducted (AC) speech with bone‐conducted (BC) speech for the purpose of utilizing them in packet loss concealment (PLC) for speech quality of voice over internet protocol (VoIP). Instead of the autocorrelation method of linear prediction (LP), which was utilized in the conventional PLC techniques, we employ the modified covariance (MC) method. The MC method provides accurate LP estimation from short input data samples and avoids the numerical problem the autocorrelation method suffers from. The lost frame is compensated from both forward and backward directions in which linear gain and weighting are applied. When BC speech is used as the input speech data for PLC in the case where the speech sender is in noisy environments, BC speech behaves more accurately than the corresponding AC speech, resulting in an excellent performance of the LP‐based PLC technique. This unveils a useful use of BC speech in speech information systems. Experiments show that in severe noise environments of AC speech‐to‐noise ratio being less than 10 dB, BC speech is superior to AC speech for PLC. It is also shown that the transmitted BC speech is more accurately reconstructed for PLC than transmitted AC speech is done. © 2023 The Authors. IEEJ Transactions on Electrical and Electronic Engineering published by Institute of Electrical Engineer of Japan and Wiley Periodicals LLC.

Real-Time milling tool breakage monitoring based on multiscale standard deviation diversity entropy

Tool breakage monitoring (TBM) during milling is vital for ensuring product quality, equipment and operator safety. This paper proposes a novel real-time TBM method based on multiscale standard deviation diversity entropy (MSDDE). Unlike existing similar works that only validate the effectiveness of the extracted features while ignoring the requirements of real-time detection, the proposed method can extract effective features from short data samples containing only two spindle rotation periods and achieve computational efficiency in milliseconds. Consequently, it is more suitable for application in the real-time detection of milling tool breakage. MSDDE quantifies the dynamic complexity of vibration signals at different time scales, which is capable of reducing the requirement for the number of spindle periods in the sliding time window while comprehensively describing the tool breakage characteristics. The proposed algorithm exhibits a time complexity of O(MN) where M depends on the scale parameter and embedding dimension. Experimental investigations show that the proposed method is able to complete the calculation of feature values within two milliseconds and realize real-time detection of milling tool breakage.

A shot for the US economy

While previous literature examines the effects of increasing COVID-19 incidences and fatality rates on economic activity, the impact of vaccination roll-outs on public health and the economy is not yet well understood. We examine the effect of a vaccination shock in the United States on various pandemic and economic indicators. By employing a BVAR model to overcome the short data sample, we show that an increase in vaccinations is not only associated with declining incidences, reproduction and fatality rates, but also increases mobility, which dampens the effect on public health indicators in the medium term. With respect to the economy, a vaccination shock is associated with lower unemployment, higher GDP growth and also reduces uncertainty in financial markets.

A New Unconstrained Approach to GNSS Atmospheric Water Vapor Tomography

AbstractA new atmospheric tomographic model totally based on Global Navigation Satellite System (GNSS) observations is proposed and tested against field observations. The method does not require a first guess, does not contain specific constraints on the variability of water vapor density inside the tomographic domain, and is able to produce reasonable results at 6 km horizontal and 500 m vertical resolutions, from short (30 min) GNSS data samples. The inversion method uses the Moore‐Penrose pseudoinverse, which is made possible by increasing the rank of the design matrix through angular interpolation and extrapolation of the observations. Comparisons against 30 consecutive 4‐h radiosonde observations and model simulations suggest the ability of the method to detect inversions and local maxima aloft, and behave sensibly in the far‐field. Further improvements from this method may be expected from higher density and multi‐constellation networks.

Natural Frequencies Identification by FEM Applied to a 2-DOF Planar Robot and Its Validation Using MUSIC Algorithm.

In this paper, the natural frequencies (NFs) identification by finite element method (FEM) is applied to a two degrees-of-freedom (2-DOF) planar robot, and its validation through a novel experimental methodology, the Multiple Signal Classification (MUSIC) algorithm, is presented. The experimental platforms are two different 2-DOF planar robots with different materials for the links and different types of actuators. The FEM is carried out using ANSYS™ software for the experiments, with vibration signal analysis by MUSIC algorithm. The advantages of the MUSIC algorithm against the commonly used fast Fourier transform (FFT) method are also presented for a synthetic signal contaminated by three different noise levels. The analytical and experimental results show that the proposed methodology identifies the NFs of a high-resolution robot even when they are very closed and when the signal is embedded in high-level noise. Furthermore, the results show that the proposed methodology can obtain a high-frequency resolution with a short sample data set. Identifying the NFs of robots is useful for avoiding such frequencies in the path planning and in the selection of controller gains that establish the bandwidth.

Input Doubling Method based on SVR with RBF kernel in Clinical Practice: Focus on Small Data

Abstract In recent years, machine-learning-based approaches have become of considerable interest to the efficient processing of short or limited data samples. Its so-called small data approach. This is due to the significant growth of new intellectual analysis tasks in various industries, which are characterized by limited historical data. These include Materials Science, Economics, Medicine, and so on. An effective processing of short datasets is especially acute in medicine. Insufficient number of vectors, significant gaps in the data collected during the supervision of patient’s treatment or rehabilitation, reduces the effectiveness or prevents effective intellectual analysis based on them. This paper presents a new approach to processing short medical data samples. The basis of the developed method is SVR with RBF kernel. The algorithmic implementation of the method in both operation modes is described. Experimental modeling on a real short data set (Trabecular bone data) is conducted. It contained only 35 observations. A comparison of the method with a number of existing machine learning methods is conducted. It is experimental established the highest accuracy of the method among those considered. The developed method has potential opportunities for wide application in various fields of medicine.

Multilayer GMDH-neuro-fuzzy network based on extended neo-fuzzy neurons and its application in online facial expression recognition

Real-time image recognition is required in many important practical problems. Interaction with users in online mode requires flexibility and adaptability from applications. The Group Method of Data Handling (GMDH) allows changing the model structure and adjusting the system architecture to the characteristics of each task under consideration. Moreover, the approximating properties of neo-fuzzy neurons used as elements of the system provide the high recognition accuracy under conditions of short data samples. This paper proposes a multilayer GMDH-neuro-fuzzy network based on extended neo-fuzzy neurons. The learning algorithm has filtering and tracking properties, guarantees the required speed important for real-time applications. The effectiveness of the proposed system is confirmed for the human emotions recognition.

Нейроподібна структура для задач прогнозування в умовах коротких вибірок даних

Задача прогнозування є однією із пріоритетних задач бізнесу. З ростом інструментарію прогнозування, а також із бурхливим розвитком потужностей комп'ютерної техніки цій задачі приділяють дедалі більше уваги. На сьогодні більшість менеджерів провідних компаній мають змогу застосовувати інформаційні системи на основі складного математичного апарату для аналізу даних. Проте проблема розуміння алгоритмів, закладених в основі таких інформаційних систем, а також правильний підбір моделі прогнозування є важливою проблемою, оскільки некоректні прогнози можуть призвести до прийняття неправильного рішення. Проблема поглиблюється у разі опрацювання недостатньої кількості даних, що характерно для розв'язання низки задач. Зокрема, для розв'язання задач прогнозування попиту на новий товар чи нову послугу організації, зокрема в системах електронної комерції, необхідна достатня кількість даних для реалізації процедур навчання. Проте їх невелика кількість, під час застосування наявних методів, призводить до неточних, некоректних прогнозів. Саме тому виникає потребу удосконалення наявних та пошуку нових рішень розв'язання задачі прогнозування в умовах коротких вибірок даних. У роботі запропоновано новий, розроблений авторами, інструмент обчислювального інтелекту для ефективного розв'язання цієї задачі. Описано нейроподібну структуру для підвищення точності розв'язання задач прогнозування в умовах коротких вибірок даних. ЇЇ побудовано з використанням штучної нейронної мережі узагальненої регресії та гібридної нейроподібної структури Моделі Послідовних Геометричних Перетворень з RBF ядром. Подано алгоритмічну реалізацію побудови методу подвоєння входів, що знаходиться в основі роботи системи. Проаналізовано процедури підбору оптимальних параметрів для її роботи. Проведено експериментальне моделювання роботи нейроподібної структури для розв'язання задачі прогнозування. Отримані результати показали високу точність її роботи. Ефективність використання розробленої структури підтверджено порівнянням її роботи з наявними – багатошаровим перцептроном, штучною нейронною мережею узагальненої регресії та машиною опорних векторів. Розроблена нейроподібна структура демонструє точність на основі MAPE відповідно у більше ніж 3, 6 та 10 разів вищу порівняно із відомими методами. Розроблену структуру можна використовувати у багатьох сферах, зокрема електронній комерції, бізнес-аналітиці, тощо.

Structural VARs and the Monetary Transmission Mechanism in Low-Income African Countries†

AbstractStructural Vector Autoregression (SVAR) methods suggest the monetary transmission mechanism may be weak and unreliable in many low-income African countries. But are structural VARs identified via short-run restrictions capable of detecting a transmission mechanism when one exists, under research conditions typical of low-income countries (LICs)? Using a small DSGE as our data-generating process, we assess the impact on VAR-based inference of short data samples, measurement error, high-frequency supply shocks, and other features of the LIC environment. The impact of these features on finite-sample bias appears to be relatively modest when identification is valid—a strong caveat, especially in LICs. Nonetheless many of these features undermine the precision of estimated impulse responses to monetary policy shocks, and cumulatively they suggest that statistically and economically insignificant results can be expected even when the underlying transmission mechanism is strong. These data features not only undermine the efficacy of the SVAR methodology for research and policy-making, but are also severe enough to motivate a continued search for monetary policy rules that are robust to these limitations.

Improving extreme wind speed prediction based on a short data sample, using a highly correlated long data sample

This paper focuses on the problem of improving extreme wind speed estimates for locations with only a short time series of measured values available, given the opportunity to use a highly correlated long time series of wind speed data. For this purpose, an efficient transfer of information is necessary between two highly correlated stochastic processes. To illustrate the efficiency of the proposed technique, two sets of two highly correlated time series of wind speed data are used. The first set of data consists of two synthetic wind speed data sets. The second set of data was obtained from two Norwegian wind speed measurement stations. This paper intends to stimulate further research and fill a gap in the current literature on this important topic.

A Simplified Methodology to Evaluate the Design Specifications of Hydraulic Components

The fatigue of a hydraulic component inherently varies due to various factors that can be divided into two categories: structural and load spectrum variability. The effects of both variabilities must be considered when determining fatigue life. Compared with the structural variability, determining the variability in the load spectrums is more difficult because the service conditions are complicated and the measurements of the load parameters are slow and expensive. The problem that arises when studying the fatigue behaviour of such components is the transferability of short data samples from real-life load histories, which are application-dependant, to laboratory test methods. Derived from the experimental background and know-how of the authors, this paper proposes a methodology that allows the definition and establishment of the hydraulic cylinder design specificactions, while taking into account the probabilistic characterisation of the load spectrum variability. This methodology could be extrapolated to other hydraulic or mechanical components.

Long-Length Critical Current Measurement of MgB2 Wire in a Coil

MgB <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> opens the door to viable, cryogen-free, and thermally stable MRI magnets. A multiturn multilayer MgB <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> solenoid coil was tested in a variable temperature environment from 5 to 34 K in order to validate the conductor's long-length critical current and n-value performance. The reacted wire was successfully wound and impregnated without imparting noticeable strain damage to the MgB <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> filaments as demonstrated by the consistency between measured long-length wire performance of the coil and the short sample data. Expected coil performance was calculated by taking into account ferromagnetic wire matrix and critical current scaling based on current percolation model.

VAR meets DSGE: Uncovering the Monetary Transmission Mechanism in Low-Income Countries

VAR methods suggest that the monetary transmission mechanism may be weak and unreliable in low-income countries (LICs). But are structural VARs identified via short-run restrictions capable of detecting a transmission mechanism when one exists, under research conditions typical of these countries? Using small DSGEs as data-generating processes, we assess the impact on VAR-based inference of short data samples, measurement error, high-frequency supply shocks, and other features of the LIC environment. The impact of these features on finite-sample bias appears to be relatively modest when identification is valid—a strong caveat, especially in LICs. However, many of these features undermine the precision of estimated impulse responses to monetary policy shocks, and cumulatively they suggest that “insignificant” results can be expected even when the underlying transmission mechanism is strong.

VAR Meets DSGE: Uncovering the Monetary Transmission Mechanism in Low-Income Countries

VAR methods suggest that the monetary transmission mechanism is weak and unreliable in low-income countries. But are structural VARs identified via short-run restrictions capable of detecting a transmission mechanism when one exists, under research conditions typical of these countries? Using small DSGEs as data-generating processes, we assess the impact on VAR-based inference of short data samples, macroeconomic volatility, high-frequency supply shocks, and other features of the LIC environment. Many of these features undermine the precision of estimated impulse responses to monetary policy shocks, but their impact on finite-sample bias appears to be relatively modest when identification is valid.

Seasonal and cyclical behaviour of farm gate milk prices

Purpose – The purpose of this paper is to analyze how the German, Irish and average EU farm gate milk prices have changed after the common agricultural policy (CAP) reform in 2003. In addition the dynamics of these prices are compared to a US farm gate price. Design/methodology/approach – These milk price time series are divided into two time periods, pre and post the CAP 2003 reform, and decomposed into their trend, seasonal and cyclical components. For the decomposition a state space model is used following the approach of Harvey (1989). Findings – The results show that the dynamics of the EU, German and Irish series converged after the CAP 2003 reform were implemented and that a three-year cycle is underlying the European milk prices which is comparable with the cycle length of the US milk price. In addition it is shown that most of the observed price variation in recent times is attributed to the cyclical component. Research limitations/implications – The division of the milk price time series into periods pre and post the CAP 2003 reform is somewhat subjective because not all measures were immediately applied after the reform. It is also possible that other factors may have contributed to the changed dynamics which have been observed. In addition this leads to a short data sample. Practical implications – The results show that policy makers should consider counter cyclical policy measures given the importance of the cyclical component. Also most models used to evaluate policies do not account for cycles which may lead to wrong conclusions. In addition farmer should be aware of the cyclical nature of milk prices as they budget and plan for the future. Originality/value – No previous decomposition studies of European milk prices exist.

Tail events in the FX markets since 1740

Purpose – The purpose of this paper is to explore the extent of the so-called “small-sample problem” within quantitative exchange-rate risk management. Design/methodology/approach – The authors take a closer look at the frequency distribution of nominal price changes in the European foreign exchange markets. Findings – The analysis clearly illustrates the risk of seriously underestimating the probability and magnitude of tail events when frequency distributions are derived from fairly short data samples. Practical implications – The authors suggest that financial institutions and regulators should have an eye for the long-term historical perspective when designing sensitivity tests or “worst case” scenarios in relation to risk assessments and stress tests. Originality/value – The authors add to the literature by analysing the distribution of nominal exchange-rate fluctuations on the basis of a unique quarterly data set for ten European exchange-rate pairs covering a time span of 273 years constructed by the authors. To the best of the authors' knowledge this is the first study on nominal exchange-rate changes for a large number of exchange-rate pairs based on quarterly data spanning almost three centuries.

Uncertainty compliant design flood estimation

Hydraulic infrastructures are commonly designed with reference to target values of flood peak, estimated using probabilistic techniques, such as flood frequency analysis. The application of these techniques underlies levels of uncertainty, which are sometimes quantified but normally not accounted for explicitly in the decision regarding design discharges. The present approach aims at defining a procedure which enables the definition of Uncertainty Compliant Design (UNCODE) values of flood peaks. To pursue this goal, we first demonstrate the equivalence of the Standard design based on the return period and the cost-benefit procedure, when linear cost and damage functions are used. We then use this result to assign an expected cost to estimation errors, thus setting a framework to obtain a design flood estimator which minimizes the total expected cost. This procedure properly accounts for the uncertainty which is inherent in the frequency curve estimation. Applications of the UNCODE procedure to real cases leads to remarkable displacement of the design flood from the Standard values. UNCODE estimates are systematically larger than the Standard ones, with substantial differences (up to 55%) when large return periods or short data samples are considered.

A review of conductor performance for the LARP high-gradient quadrupole magnets

We summarize critical current measurements and parameterizations of the data of 112 round wires and extracted strands that were reacted with the first 17 coils for the high-gradient quadrupole (HQ) magnets for the US LHC Accelerator Research Program (LARP). We standardize the strand parameterizations and coil ‘short sample’ calculations, and demonstrate that the entire critical current database can be captured in two scaling parameters per coil. These parameters summarize the short sample performance for each coil for either HQ magnet tests, or mirror tests of individual coils. We also demonstrate that for RRP® conductors, generic strain scaling parameters can be derived for at least four substantially different wire configurations, and standardize self-field corrections for LARP. The parameterized conductor performance is used to judge the performance of the HQ magnets and mirror tests. We find that although the HQ magnets reach around 86% of their short sample limitations, they are limited by factors other than the critical current of the conductor. Individual coils in mirror tests reach up to 98% of the expected performance, and do appear limited by the critical current of the conductor. Detailed analysis of short sample performance through accurate parameterizations simplifies the accessibility of short sample data, and enables accurate judgment of magnet performance as well as conductor and cable quality.

Application of qualitative analysis techniques to the environmental modeling of plant species cultivation

The main issue discussed in the paper is the application of qualitative analysis techniques when short data samples are available at the initial stage of experiments carried out to study behaviors of agricultural plant species under different environmental conditions. The proposed techniques are described on the examples of two tasks: analyze behaviors of chemical elements in plant species irrigated with treated wastewater (on the example of Brocoli cabbage), as well as in plant species cultivated near the highway (on the example of Lycopersicum species). The system of ecological interactions is decomposed into individual subsystems with sufficiently small number of variables describing concentrations of elements. The Group Method of Data Handling (GMDH) is applied to obtain the regressions between variables within subsystems. The Qualitative Description based on signs combinations of differences for a regression function is employed to qualitatively analyze the relationships among variables, both within individual subsystems and within the system in whole. This approach makes it possible to identify main behavior features of individual objects/relationships between objects, classify behaviors according to different criteria of similarity/distinction, provide feature matching, etc. The obtained results then can be used to design the next-stage experiments and construct more comprehensive models of plant species behaviors.