Sampled Time Series Research Articles

Metabarcoding revealed a high diversity of Amphidomataceae (Dinophyceae) and the seasonal distribution of their toxigenic species in the Taiwan Strait

The dinophyte family Amphidomataceae includes the genera Azadinium and Amphidoma. Four of these species are known to produce azaspiracids, which are lipophilic phycotoxins accumulating in shellfish. The diversity and biogeography of Amphidomataceae is far from yet resolved. Here we performed a time series sampling of both water and sediments in the Taiwan Strait from Nov. 2018 to April 2021. Metabarcoding was performed to unveil the diversity of Amphidomataceae targeting internal transcribed spacer (ITS1) region and partial large subunit ribosomal DNA (LSU rDNA D1–D3), followed by quantitative PCR (qPCR) with modified primers for Az. poporum ribotypes. The diversity of Amphidomataceae was revealed from the water samples with the aid of ITS1 and LSU based molecular phylogeny. The LSU based approach detected only a few species. In contrast, ITS1 based dataset showed eight new Azadinium clades and several ZOTUs (zero-radius operational taxonomic units) grouping together with Am. languida. Moreover, eleven known Azadinium species including three ribotypes of Az. poporum and Az. dexteroporum, and two ribotypes of Az. spinosum, were detected. The latter two species have not been reported in China before. Among these toxigenic species, Az. poporum was relevantly abundant whereas others were rare. The maximum of 209 cells L −1 of Az. poporum ribotype A was estimated using qPCR nearby Quanzhou in Nov. 2018 and 172 cells L 1 of Az. poporum ribotype B was detected far off coast in Apr. 2021. Metabarcoding on sediment samples revealed Az. poporum ribotypes B and C, but strains obtained with sediment incubation experiments yielded only ribotype B. Using qPCR about 0.2 cysts g −1 of Az. poporum ribotype B were quantified in May 2019 but cysts of Az. poporum ribotype C were not detected. Our results suggest that metabarcoding targeting ITS1 region is powerful to uncover the diversity of harmful dinophytes. Our results also highlight the rich diversity of Amphidomataceae and risk potential of azaspiracids in the Taiwan Strait and surrounding waters.

Stochastic interpolation of sparsely sampled time series by a superstatistical random process and its synthesis in Fourier and wavelet space

We present a novel method for stochastic interpolation of sparsely sampled time signals based on a superstatistical random process generated from a multivariate Gaussian scale mixture. In comparison to other stochastic interpolation methods such as Gaussian process regression, our method possesses strong multifractal properties and is thus applicable to a broad range of real-world time series, e.g. from solar wind or atmospheric turbulence. Furthermore, we provide a sampling algorithm in terms of a mixing procedure that consists of generating a -dimensional field , where each Gaussian component is synthesized with identical underlying noise but different covariance function parameterized by a log-normally distributed parameter ξ. Due to the Gaussianity of each component , we can exploit standard sampling algorithms such as Fourier or wavelet methods and, most importantly, methods to constrain the process on the sparse measurement points. The scale mixture u(t) is then initialized by assigning each point in time t a and therefore a specific value from , where the time-dependent parameter follows a log-normal process with a large correlation time scale compared to the correlation time of . We juxtapose Fourier and wavelet methods and show that a multiwavelet-based hierarchical approximation of the interpolating paths, which produce a sparse covariance structure, provide an adequate method to locally interpolate large and sparse datasets.

ANALISIS REGRESI FAKTOR PANEL DINAMIS BLUNDELL-BOND DENGAN ESTIMASI SYSTEM-GENERALIZED METHOD OF MOMENT PADA SAHAM FARMASI DI BEI

The pharmaceutical sector has become a concern during the Covid-19 pandemic because of the large use of drugs. Companies need to improve financial performance to increase their share prices and investors need analysis to predict future stock prices. This study aims to analyze the influence of stock prices on 10 pharmaceutical companies on the Indonesia Stock Exchange during the third quarter of 2020 to the third quarter of 2021. Based on previous research, the factors that are thought to have an effect on changes in stock prices are internal financial ratios (ROA, ROE, NPM, GPM, EPS, PER, BV, PBV, DAR, DER, CR, QR, Cash Asset Ratio) and external inflation, exchange rates, interest rates. The method used in this research is dynamic panel factor regression analysis with GMM (Generalized Method of Moment) estimation. Factor analysis to reduce the independent variables to form a factor score which is then entered into the regression. The regression model was obtained from the comparison of Arellano-Bond GMM and Blundell-Bond System. The GMM system is the development of Arellano-Bond which will produce more efficient estimates when the sample time series is short. The results of the study were obtained 3 factor scores with a total variance of 81.757% from the elimination of 6 variables that had MSA <0.5. The best model is the Blundell-Bond Twostep System which fulfills the model assumptions with RMSE 803.276.

Groundwater discharge and streams drive spatial alkalinity and pCO2 dynamics in two contrasting tropical lagoons

Coral reef lagoons are areas of complex carbon cycling, however, regional (e.g. land use) and global (e.g. climate) factors, including land runoff and ocean acidification, are adversely affecting carbonate-building coral reef systems. Coupled with this, surplus nutrients entering coastal waters can prompt excess algae growth, which can stimulate further carbon dioxide (CO2) production in the water column, thus enhancing coral/sediment dissolution. However, new inputs of alkalinity into coastal systems can buffer against acidification. By combining the natural groundwater tracer radon (222Rn) with carbonate chemistry in two contrasting Cook Island lagoons (the fringing Muri Lagoon on Rarotonga and the comparatively larger Aitutaki near-atoll lagoon), we were able to identify multiple drivers of coral reef acidification and regulation. Despite the lagoons having similar rates of submarine groundwater discharge (3.1 to 3.3 cm d−1; although the rate in Aitutaki is a maximum rate based on an assumed 100 m seepage face), groundwater inputs of CO2 and alkalinity were primarily driven by different sources (discrete offshore seeps in Muri Lagoon and dredged channels in Aitutaki Lagoon). Streams delivered low alkalinity water to both lagoons, but high pCO2 waters to the Aitutaki Lagoon in contrast to Muri Lagoon. Aragonite saturation states (ΩAr) ranged between 2.2 and 5.2, with areas of low ΩAr corresponding to areas of high radon and excess algal growth in Muri Lagoon, and areas that receive low alkalinity surface water in Aitutaki. Time-series sampling indicated that tidal heights and the ability of seawater to overtop the fringing reef influenced groundwater dynamics, lagoon hydrodynamics and carbonate chemistry. Groundwater discharge and stream flows were a significant freshwater source of new geologic CO2 and alkalinity to each lagoon, while recirculated seawater is likely a significant source of biologic CO2 driven by microbial respiration in sediments. The study found that while groundwater inputs of alkalinity may reduce acidification, they do not fully counteract ongoing acidification and CO2 inputs. This study also highlighted the need for future studies to undertake detailed spatial measurements to accurately characterise tropical island carbon dynamics due to the heterogeneous nature of these environments.

A Flexible Monitoring Framework via Dynamic-Multilayer Graph Convolution Network

Due to the continuous technological innovation in industrial processes, many deep learning based methods have shown powerful capability in handing equipment status monitoring, but most of them ignore the temporal features and the dynamic changes of diverse spatial structure of the raw data. Meanwhile, these methods usually focus on handling a single downstream task, but rarely consider different tasks simultaneously. To solve these issues, this paper proposes a more flexible monitoring framework based on dynamic-multilayer graph convolution network, which can be adapted to different downstream tasks simultaneously by agile combinations of the modules according to the different industrial scenarios. First, the time series of fault samples are segmented and constructed into a feature matrix to extract the temporal information by the temporal module. At the same time, in order to fully characterize the dynamic change of different spatial structures among samples, the dynamic graph of each moment is expanded into a multilayer graph through various composition indexes. Then, the spatial information is extracted by the intra-layer and inter-layer convolution operations in the spatial module, and the fused features are applied to different downstream tasks. Finally, the experiments are performed in two different downstream tasks, namely three datasets for fault diagnosis and one dataset for remaining useful life prediction. The results of both diagnosis and prediction are better than those of the comparison algorithms, so the effectiveness and flexibility of the proposed framework are proved.

An Efficient Sequential Embedding ConvNet for Rotating Machinery Intelligent Fault Diagnosis

Effective monitoring of critical components of rotating machinery can bring considerable economic benefits and ensure personal safety. However, effective information in mechanical vibration signals is usually submerged by noise interference and non-stationary operating conditions. This causes constructed time-series vibration samples to produce temporal covariate shift (TCS) phenomenon, resulting in differences in the distribution of the training and testing sets. Focusing on this issue, this study proposes an efficient sequential embedding ConvNet (SECN), which can be mainly divided into three layers. Sequential embedding (SE) is firstly proposed to reduce the distribution differences between samples by trainable linear dimensional boosting, where SE aims to realize the secondary representation of samples information distribution via the expansion of a low-dimensional space to a high-dimensional space. Then a single layer of multiscale separable convolution is introduced to extract information over different perceptual fields. Finally the fully connected layer is synthesized by global average pooling to complete the classification task. Experiments and comparisons on three datasets verified that SECN has obvious advantages in terms of accuracy, noise immunity and sample number dependence. The key principles of the proposed SE are explored by ablation and quantitative experiments. The SE applied in the proposed SECN model shows great potential as a basic framework for intelligent fault diagnosis models.

The use of lichen and fungal collections in Australian herbaria to identify temporal changes in air pollution and assist environmental management: challenges to conserving herbarium specimens for the future

Context Global environmental modification is placing increased demand on taxonomic resources, especially herbarium collections, which are being utilised for information and as biological resource material in ways never previously envisaged. This increased use of herbarium collections presents current and on-going management challenges. Aims To demonstrate, with two Australian studies, the essential role of herbarium collections in investigating long-term pollution; to emphasise the importance of on-going collecting for herbaria and make recommendations for future collection management. Methods Australian case studies of a long-term pollution problem are used to show how verified lichen and fungal taxa in Australian herbaria were crucial in identifying sources of air pollution over time. Key results Chemical and isotopic analyses of lead in samples from curated herbarium collections were used to reconstruct patterns of atmospheric lead deposition over 150 years in eastern Australia. This extended time series of samples facilitated documentation of the full cycle from natural background concentrations to the introduction and increasing usage of lead both in industry and motor vehicles, to the eventual discontinuation of its use in automotive fuel. In addition to demonstrating the value of regular, long-term sampling of selected taxa and potential applications to emerging environmental problems, the importance of strict protocols for destructive sampling and analysis of herbarium specimens, especially for older collections containing small quantities of material, has been highlighted. Conclusions Both the Australian and previous studies have identified a number of problems and potential challenges to maintaining and maximising the utilisation of finite herbarium resources for the future. While acknowledging that Australasian herbaria have well documented policies and protocols for studies that sample their collections, six recommendations are made, for future implementation as appropriate. Implications Current management strategies for finite reference collections are summarised and recommendations for the future include the following: repeated, ongoing collecting over broader areas; periodic re-collection of larger samples from select reference sites with detailed field data; preparation and storage of specimens under controlled environmental conditions, with minimal chemical pest control; utilisation of common widespread taxa for research; careful supervision of destructive sampling techniques to assess compounds present; and adequate funding to facilitate the digitisation of herbarium specimens to increase access and optimise use of curated collections.

A Novel Deep Parallel Time-Series Relation Network for Fault Diagnosis

Currently, deep learning based methods are widely used in the fault diagnosis of time series data for their high precision. However, the application of traditional deep learning fault diagnosis models is limited by their calculational efficiency and poor interpretation ability. To address the problems, a fault diagnosis model named the deep parallel time-series relation network (DPTRN) is proposed in this paper. There are three main advantages of the DPTRN. (1) Our proposed time relationship unit can perform feature extraction on each time node of a time series sample simultaneously, therefore, the DPTRN performs fault diagnosis in a parallel way and improves the computing efficiency significantly. (2) By improving the absolute position embedding, our novel decoupling position embedding unit can be directly applied for fault diagnosis and can learn contextual information. (3) Our proposed DPTRN has an obvious advantage in feature interpretability compared with traditional deep learning based models. Applying the DPTRN model on four datasets, we achieved higher diagnosis performance with much lower cost, which indicates the effectiveness, efficiency and interpretability of the proposed DPTRN model.

Electricity price forecast based on the STL-TCN-NBEATS model

Taking long-term high-frequency electricity price data as the research content, this paper proposes seasonal and trend decomposition using loess-temporal convolutional network-neural basis expansion analysis for an interpretable time series forecasting (STL-TCN-NBEATS) model to solve the problems of low forecast accuracy caused by high volatility, high frequency and nonlinearity and poor interpretability of the deep learning model. By comparing the forecast effects of the temporal convolutional network-long short-term memory (TCN-LSTM), LSTM and other models, the main conclusions are as follows: (1) The hybrid model, STL-TCN-NBEATS, selected in this paper can effectively solve the problem of low forecast accuracy after reasonable selection of model parameters. The evaluation indexes of the root mean square error (RMSE) and the mean absolute percentage error (MAPE) were 3.7441 and 4.5044, respectively, which were 3.1416 and 2.1336 lower than those of the second-best model (TCN-LSTM). Compared with the autoregressive integrated moving average model (ARIMA), the accuracy was improved by approximately 49.18% (RMSE) and 60.35% (MAPE). (2) The STL-TCN-NBEATS model has better feature extraction ability, so it can obtain higher forecast accuracy. Since the construction of the TCN introduces extended causal convolution and residual blocks, the deep learning network model has better processing ability and robustness for large sample time series. Moreover, the NBEATS network structure enables the model to be trained quickly, and the experimental results verify the effectiveness and high accuracy of this method. (3) The model not only has high precision but also has some interpretability. By decomposing time series data into a trend term, period term and remainder term, the NBEATS and the TCN are used to process the trend term, period term and remainder term, respectively, so that the hybrid model can forecast electricity prices according to the traditional time series processing mode.

Corruption and FDI Inflows in Indonesia

The inflows of Foreign Direct Investment (FDI) to a country depend on many factorszzne of the crucial factors is the corruption index, which represents a pivotal aspect in Global Investment Competitiveness (GIC). This research analyzes the impact of the corruption perception index on FDI inflows to Indonesia. Other variables affecting FDI are Gross Domestic Product (GDP), inflation rate, and population. Vector Error Correction Model (VECM) is used to estimate the time series sample between 1995 and 2019. Following the time series procedure, several tests are conducted on the dataset before applying VECM: unit-root, optimum lag, and cointegration. The main finding shows that the higher the corruption index, the higher the FDI Inflows, implicating that low corruption attracts more FDI inflows to Indonesia. Other findings are that an increase in GDP attracts more FDI, an increase in inflation rate reduces FDI inflows, and an increase in population increases FDI inflows. The policy implication of the findings is related to an effort in attracting more FDI inflows requires a substantial endeavor in combating corruption and increasing the Corruption Perception Index of Indonesia from time to time.

Selective extracellular secretion of small double-stranded RNA by Tetragenococcus halophilus.

Small double-stranded RNAs (dsRNAs) abundantly produced by lactic acid bacteria demonstrate immunomodulatory activity and antiviral protective immunity. However, the extracellular secretion of dsRNA from lactic acid bacteria and their compositional and functional differences compared to the intracellular dsRNA is unknown. In this study, we compared the intracellular and secreted extracellular dsRNA of the lactic acid bacteria, Tetragenococcus halophilus, commonly present in fermented foods, by growing in RNA-free and RNase-free media. We used RNA deep sequencing and in-silico analysis to annotate potential regulatory functions for the comparison. A time series sampling of T. halophilus culture demonstrated growth phase-dependent dynamics in extracellular dsRNA secretion with no major change in the intracellular dsRNA profile. The RNA deep sequencing resulted in thousands of diverse dsRNA fragments with 14-21 nucleotides in size from T. halophilus culture. Over 70% of the secreted extracellular dsRNAs were unique in their sequences compared to the intracellular dsRNAs. Furthermore, the extracellular dsRNA abundantly contains sequences that are not T. halophilus genome encoded, not detected intracellularly and showed higher hits on human transcriptome during in-silico analysis, which suggests the presence of extrachromosomal mobile regulatory elements. Further analysis showed significant enrichment of dsRNA target genes of human transcriptome on cancer pathways and transcription process, indicating the extracellular dsRNA of T. halophilus is different not only at the sequence level but also in function. Studying the bacterial extracellular dsRNA is a promising area of future research, particularly for developing postbiotic fermented functional foods and understanding the impact of commensal gut bacteria on human health.

Magma storage and transport timescales for the 1959 Kīlauea Iki eruption and implications for diffusion chronometry studies using time-series samples versus tephra deposits

Magma storage and transport timescales for the 1959 Kīlauea Iki eruption and implications for diffusion chronometry studies using time-series samples versus tephra deposits

Global land surface 250 m 8 d fraction of absorbed photosynthetically active radiation (FAPAR) product from 2000 to 2021

Abstract. The fraction of absorbed photosynthetically active radiation (FAPAR) is a critical land surface variable for carbon cycle modeling and ecological monitoring. Several global FAPAR products have been released and have become widely used; however, spatiotemporal inconsistency remains a large issue for the current products, and their spatial resolutions and accuracies can hardly meet the user requirements. An effective solution to improve the spatiotemporal continuity and accuracy of FAPAR products is to take better advantage of the temporal information in the satellite data using deep learning approaches. In this study, the latest version (V6) of the FAPAR product with a 250 m resolution was generated from Moderate Resolution Imaging Spectroradiometer (MODIS) surface reflectance data and other information, as part of the Global LAnd Surface Satellite (GLASS) product suite. In addition, it was aggregated to multiple coarser resolutions (up to 0.25∘ and monthly). Three existing global FAPAR products (MODIS Collection 6; GLASS V5; and PRoject for On-Board Autonomy–Vegetation, PROBA-V, V1) were used to generate the time-series training samples, which were used to develop a bidirectional long short-term memory (Bi-LSTM) model. Direct validation using high-resolution FAPAR maps from the Validation of Land European Remote sensing Instrument (VALERI) and ImagineS networks revealed that the GLASS V6 FAPAR product has a higher accuracy than PROBA-V, MODIS, and GLASS V5, with an R2 value of 0.80 and root-mean-square errors (RMSEs) of 0.10–0.11 at the 250 m, 500 m, and 3 km scales, and a higher percentage (72 %) of retrievals for meeting the accuracy requirement of 0.1. Global spatial evaluation and temporal comparison at the AmeriFlux and National Ecological Observatory Network (NEON) sites revealed that the GLASS V6 FAPAR has a greater spatiotemporal continuity and reflects the variations in the vegetation better than the GLASS V5 FAPAR. The higher quality of the GLASS V6 FAPAR is attributed to the ability of the Bi-LSTM model, which involves high-quality training samples and combines the strengths of the existing FAPAR products, as well as the temporal and spectral information from the MODIS surface reflectance data and other information. The 250 m 8 d GLASS V6 FAPAR product for 2020 is freely available at https://doi.org/10.5281/zenodo.6405564 and https://doi.org/10.5281/zenodo.6430925 (Ma, 2022a, b) as well as at the University of Maryland for 2000–2021 (http://glass.umd.edu/FAPAR/MODIS/250m, last access 1 November 2022).

Design and Optimization of Multipoint Sampler for Seafloor Sediment Carried by a Deep-Sea Landing Vehicle

The present study proposes a low-energy consumption multipoint sampler carried by a deep-sea landing vehicle (DSLV) to meet the requirements of time series sampling in local areas and location series sampling in wide areas, and an optimization method of sampling structure based on least-squares support-vector machine (LSSVM) surrogate model and a multi-objective particle swarm optimization (MOPSO) algorithm. First, the overall structure and core components, such as the multipoint sampler’s sampling structure, were designed. The optimization variables were the cone angle, sampling tube inner diameter, and sampling tube inner hole length, which were determined by considering the force with which the sampling structure penetrates the seafloor sediment. Then, the sampling process was simulated by the finite element method-smoothed particle hydrodynamics (FEM-SPH) method, while the accurate LSSVM model of force required for sampling and sampling tube volume was established. Finally, the MOPSO algorithm was used for multi-objective optimization of model parameters of sampling structure. The optimal model of sampling structure that can provide theoretical support for the optimal design of multipoint sampler effectively reduces energy consumption and improves sampling efficiency by force required for sampling 25.89% lower and sampling tube volume 34.81% higher than the original model.

Shapelet selection based on a genetic algorithm for remaining useful life prediction with supervised learning

RUL (remaining useful life) shapelets were recently developed to overcome the shortcomings of similarity-based RUL prediction methods, such as high sensitivity to parameters. RUL shapelets are informative subsequences whose distances to a run-to-failure time series sample are very useful for predicting the RUL of the sample. However, the prediction performance and interpretability highly depend on the set of RUL shapelets, and it is very difficult to compose an optimized set. In this paper, we mathematically formalize the RUL shapelet composition problem with multiple objective functions. In addition, we analyze the characteristics of good RUL shapelet sets and develop a solution methodology based on a genetic algorithm. From the various experiments, we validate that the proposed method outperforms previous ones and suggest how to use the proposed method. The solution methodology developed in this paper can be applied to solve various RUL prediction problems. In addition, the findings on the RUL shapelets can help researchers develop their RUL shapelet-based solution.

Machine learning-based prediction for time series damage evolution of Ni-based superalloy microstructures

The Directionally Solidified (DS) superalloy DZ125 developed time-series related microstructure damage with serve time. The conventional grey prediction model (GM (1,1)), the fractional order accumulation grey prediction model (FAGM (1,1)), and the long short-term memory (LSTM) neural network prediction model were used to forecast the values of the time-related damage variables. The damage variables of the superalloy microstructure evolution were obtained through experimental observations of the evolution characteristics of the γ' precipitates and γ matrix. The microstructure damage variables were predicted by three machine learning models, and the fatigue damage evolution mechanism of the superalloy was analyzed. It was found that the root mean square error of the three prediction models was 0.072, 0.001, and 0.008, respectively. Finally, a model to predict fatigue life was established based on the Chaboche fatigue damage theory, and the fatigue life was obtained using the damage variables predicted by machine learning. The results revealed that the fractional order cumulative grey model and the long short-term memory neural network prediction model were both more accurate, with the conventional grey model being better for exponentially small sample time series data. • Damage variable was predicted with time-series small sample using grey model. • The prediction accuracy of the GM (1,1), FAGM (1,1), and LSTM were compared. • Fatigue life prediction model of DZ125 superalloy containing damage was developed.

Seasonal changes in taxonomic, size composition, and Normalised Biomass Size Spectra (NBSS) of mesozooplankton communities in the Funka Bay, southwestern Hokkaido: Insights from ZooScan analysis

Knowledge about the taxonomic and size composition of mesozooplankton is of critical importance for both fisheries and oceanography. In this study, we collected an annual time series of mesozooplankton samples in Funka Bay and analysed them using ZooScan to reveal the seasonal changes in taxonomic and size structure. Both zooplankton abundance and biovolume peaked in April and were dominated by an appendicularian Oikopleura labradoriensis which has been reported as being are being important food source for flatfish larvae. Furthermore, Noctiluca scintillans was abundant from September to December. The occurrence of this species may be related to the recent increase in the transport of Tsugaru Warm Current water into the bay. The Normalized Biomass Size Spectra (NBSS) was significant for 16 of 21 sampling dates. Insignificant NBSS, with an extremely flat slope, was observed for January–March. Significant NBSS with a flat slope was observed in April. Then, the relatively steep slopes of the NBSS were the case from mid-May to December. These seasonal changes in the slope of NBSS suggest that the energy transfer efficiency for higher trophic levels varied seasonally. The observed slopes of the NBSS in Funka Bay ranged from −1.09 to −0.30, which was flatter than the theoretical value (−1) and the previously reported values for the oceanic region of the western North Pacific. This suggests that the energy transfer efficiency to organisms at higher trophic levels in the Funka Bay is higher than in the adjacent oceanic region.

A Multi-Model Prediction Method for Coal Mine Gas Concentration with Hierarchical Structure

ABSTRACT The low concentration of gas in the gas blending process is influenced by a number of factors and is characterized by some time variation and non-linearity. Therefore, the gas concentration needs to be predicted. This paper proposes a multi-model forecasting method with a hierarchical structure. First, because the measured gas concentration time-series data contain a lot of noise, the time-series data are decomposed into several independent eigenmode functions by using empirical mode decomposition, adaptively denoising by low-pass filtering, and then using phase space reconstruction technology to obtain a new time-series sample. Then, the training samples are grouped by conditional fuzzy clustering to determine the number of sub-modules. Finally, the maximum membership method is used to select sub-models and sub-sub-models, and then a multi-model time-series prediction model is established. The model can not only select different sub-models to process data in different regions but also can process each data jointly by multiple sub-models in different sub-models. Experiments were carried out on low-concentration measured data extracted from mines. The experimental results show that the proposed prediction model can capture the nonlinear characteristics of gas concentration time series and is superior to other existing prediction models in accuracy.

What could we have learnt from the previous flood data to predict losses caused by the 1980, 1986, and 1998 catastrophic floods in Ukrainian Transcarpathian?

This paper explores some aspects relating to retrospective predicting the confirmed monetary losses caused by the disastrous floods of 1980, 1986, and 1998 in the Tisza River basin within the Transcarpathian region of Ukraine. The research was based on two time series – the losses because of past floods and the maxima water discharges gauged at the hydrological station near the village of Vylok, Vynohradiv district. The main aim of the research was to make out whether it had been the possibility to predict the losses due to those floods in advance.In solving the task, there was revealed and modelled the dependence of the risk of losses due to the floods in Transcarpathia on the maximum water discharges of the Tisza River gauged at the “Vylok” hydrological station. Predicting was based on the hypothesis of the stationary random process for maximum water discharges, which allowed using an empirical distribution function of a random variable regarding flood water discharges assessing the risk of flood losses.Retrospective predicting of the losses caused by the floods of 1980, 1986, and 1998 was carried out by means of a combined situational-inductive predictive modelling method (CSIPMM), being an original author’s development. The method relates to predicting the behaviour of complex dynamic systems based on monitoring findings presented as time series data reflecting evolutions of a resulting (dependent) variable and an explaining (independent) variable (predictor). The method uses extrapolation-regression type models. According to this method, the prediction task is performed in two stages. The first stage realises the retrospective situational modelling task aiming to obtain a set of simple regressions (situational models) built on data of sample time series. The situational models are accepted to be adequate or relevant ones only within certain periods of time determined as situations. In the second stage, based on the generalization (on an ensemble) of the obtained retrospective situational models, inductive “levels” models are built, which reflect the behaviour of a controlled parameter of the system or process (a resulting variable) at several fixed values of a predictor in time. The inductive models are used in extrapolative predicting situational models belonging to future periods (situations).In total, three predictions were made: (1) taking into account the annual maximum flood discharges from 1954 to 1979 (before the flood of 1980); (2) the same from 1954 to 1985 (before the flood of 1986); (3) the same from 1954 to 1997 (before the flood of 1998). The study found that there had been a possibility to predict the confirmed monetary losses inflicted by the flood of 1986 and 1998 (relative predicting errors of 7.2-8.7% and 6.0-12.8% depending on the prediction options).

The new supervisory outlier test (SOT) on net interest income (NII): empirical evidence from a sample of Italian banks

This paper contributes to prior literature and to the current debate concerning the prudential supervisory framework to measure interest rate risk in the banking book (IRRBB), which has been significantly changed on April 2016, when the Basel Committee on Banking Supervision (BCBS) published the latest update of its measurement standards. The consultation launched by the European Banking Authority (EBA) on December 2021, aiming at introducing the supervisory outlier test (SOT) on net interest income (NII), presents several issues and policy implications which could influence in the next future banks' asset and liability management strategies, their internal control systems, risk policies and procedures. By analyzing a sample of 28 Italian commercial banks at the end of 2021, representing more than 70% of Italian baking system’s total assets , we observe that the thresholds proposed by the EBA appear very strict and significantly depend on: i) the sample considered, ii) the lower bound applied to interest rates in the downward scenarios and iii) the current level of interest rates term structure. Our results suggest that the proposed values should be considered with caution as it seems that their potential impacts have not been thoroughly assessed. Further analyses are therefore necessary to guarantee greater robustness of the methodology used for the calibration of the thresholds, taking also into account a wider sample of banks and longer time series, as well as the correlation between the two approaches.