2024
8.
Górecki, Tomasz; andPaweł Piasecki, Maciej Łuczak
An exhaustive comparison of distance measures in the classification of time series with 1NN method Journal Article
In: Journal of Computational Science, vol. 76, pp. 102235, 2024, ISSN: 1877-7503.
@article{Górecki2024,
title = {An exhaustive comparison of distance measures in the classification of time series with 1NN method},
author = {Tomasz Górecki and Maciej Łuczak andPaweł Piasecki},
doi = {10.1016/j.jocs.2024.102235},
issn = {1877-7503},
year = {2024},
date = {2024-02-10},
urldate = {2024-02-10},
journal = {Journal of Computational Science},
volume = {76},
pages = {102235},
abstract = {Time series classification is an important and challenging problem in data analysis. With the increase in time series data availability, hundreds of algorithms have been proposed. A huge effort over the past two decades caused a significant improvement in both the efficiency and effectiveness of time series classification. There is a belief in the community that the best method is a surprisingly simple one. Even though there exist many algorithms outperforming the nearest neighbor (NN) classifier, the popularity of the latter remains stable — due to its simplicity and high performance in many domains, especially with dynamic time warping (DTW) as the distance measure. In the paper, we present an exhaustive study in which we compare the performance of different similarity measures relying on the 1NN classifier. We used the most highly cited time series distance measures used in classification (in total we compared 56 distance measures). We evaluate methods on all datasets from the UCR Time Series Classification Archive. Additionally, we perform extensive statistical comparison of the examined methods. We show that none of the distance measures is the best for all datasets, however, there is a group performing statistically significantly better than the others},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Time series classification is an important and challenging problem in data analysis. With the increase in time series data availability, hundreds of algorithms have been proposed. A huge effort over the past two decades caused a significant improvement in both the efficiency and effectiveness of time series classification. There is a belief in the community that the best method is a surprisingly simple one. Even though there exist many algorithms outperforming the nearest neighbor (NN) classifier, the popularity of the latter remains stable — due to its simplicity and high performance in many domains, especially with dynamic time warping (DTW) as the distance measure. In the paper, we present an exhaustive study in which we compare the performance of different similarity measures relying on the 1NN classifier. We used the most highly cited time series distance measures used in classification (in total we compared 56 distance measures). We evaluate methods on all datasets from the UCR Time Series Classification Archive. Additionally, we perform extensive statistical comparison of the examined methods. We show that none of the distance measures is the best for all datasets, however, there is a group performing statistically significantly better than the others
7.
Kowal-Wisniewska, Ewelina; Jaskiewicz, Katarzyna; Bartochowska, Anna; Kiwerska, Katarzyna; Ustaszewski, Adam; Gorecki, Tomasz; Giefing, Maciej; Paluszczak, Jaroslaw; Wierzbicka, Malgorzata; Jarmuz-Szymczak, Malgorzata
Towards effectiveness of cell free DNA based liquid biopsy in head and neck squamous cell carcinoma Journal Article
In: Scientific Reports, vol. 14, pp. 2251, 2024.
@article{Kowal-Wisniewska2024,
title = {Towards effectiveness of cell free DNA based liquid biopsy in head and neck squamous cell carcinoma},
author = {Ewelina Kowal-Wisniewska and Katarzyna Jaskiewicz and Anna Bartochowska and Katarzyna Kiwerska and Adam Ustaszewski and Tomasz Gorecki and Maciej Giefing and Jaroslaw Paluszczak and Malgorzata Wierzbicka and Malgorzata Jarmuz-Szymczak },
doi = {10.1038/s41598-024-52031-5},
year = {2024},
date = {2024-01-26},
journal = {Scientific Reports},
volume = {14},
pages = {2251},
abstract = {Liquid biopsy is a minimally invasive procedure, that uses body fluids sampling to detect and characterize cancer fingerprints. It is of great potential in oncology, however there are challenges associated with the proper handling of liquid biopsy samples that need to be addressed to implement such analysis in patients’ care. Therefore, in this study we performed optimization of pre-analytical conditions and detailed characterization of cfDNA fraction (concentration, length, integrity score) in surgically treated HNSCC patients (n = 152) and healthy volunteers (n = 56). We observed significantly higher cfDNA concentration in patients compared to healthy controls (p < 0.0001) and a time dependent decrease of cfDNA concentration after tumor resection. Our results also revealed a significant increase of cfDNA concentration with age in both, healthy volunteers (p = 0.04) and HNSCC patients (p = 0.000002). Moreover, considering the multitude of HNSCC locations, we showed the lack of difference in cfDNA concentration depending on the anatomical location. Furthermore, we demonstrated a trend toward higher cfDNA length (range 35–10380 and 500–10380 bp) in the group of patients with recurrence during follow-up. In conclusion, our study provide a broad characterization of cfDNA fractions in HNSCC patients and healthy controls. These findings point to several aspects necessary to consider when implementing liquid biopsy in clinical practice including: (I) time required for epithelial regeneration to avoid falsely elevated levels of cfDNA not resulting from active cancer, (II) age-related accumulation of nucleic acids accompanied by less efficient elimination of cfDNA and (III) higher cfDNA length in patients with recurrence during follow-up, reflecting predominance of tumor necrosis.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Liquid biopsy is a minimally invasive procedure, that uses body fluids sampling to detect and characterize cancer fingerprints. It is of great potential in oncology, however there are challenges associated with the proper handling of liquid biopsy samples that need to be addressed to implement such analysis in patients’ care. Therefore, in this study we performed optimization of pre-analytical conditions and detailed characterization of cfDNA fraction (concentration, length, integrity score) in surgically treated HNSCC patients (n = 152) and healthy volunteers (n = 56). We observed significantly higher cfDNA concentration in patients compared to healthy controls (p < 0.0001) and a time dependent decrease of cfDNA concentration after tumor resection. Our results also revealed a significant increase of cfDNA concentration with age in both, healthy volunteers (p = 0.04) and HNSCC patients (p = 0.000002). Moreover, considering the multitude of HNSCC locations, we showed the lack of difference in cfDNA concentration depending on the anatomical location. Furthermore, we demonstrated a trend toward higher cfDNA length (range 35–10380 and 500–10380 bp) in the group of patients with recurrence during follow-up. In conclusion, our study provide a broad characterization of cfDNA fractions in HNSCC patients and healthy controls. These findings point to several aspects necessary to consider when implementing liquid biopsy in clinical practice including: (I) time required for epithelial regeneration to avoid falsely elevated levels of cfDNA not resulting from active cancer, (II) age-related accumulation of nucleic acids accompanied by less efficient elimination of cfDNA and (III) higher cfDNA length in patients with recurrence during follow-up, reflecting predominance of tumor necrosis.
2023
6.
Łaźniewska, Ewa; Plac, Klaudia; Górecki, Tomasz; Rykowski, Jarogniew
In: Barometr regionalny, vol. 19, no. 2, pp. 37-73, 2023.
@article{Lazniewska2023,
title = {Time to Rethink Energy: An Initiative towards Renewable and Smart Energy. A Comparative Analysis of Poland and Germany and Conclusions for Cross-Border Cooperation and Improving Resilience},
author = {Ewa Łaźniewska and Klaudia Plac and Tomasz Górecki and Jarogniew Rykowski },
doi = {10.56583/br.2292},
year = {2023},
date = {2023-12-31},
urldate = {2023-12-31},
journal = {Barometr regionalny},
volume = {19},
number = {2},
pages = {37-73},
abstract = {By analyzing the regional disparities in energy production and consumption between Poland and Germany, and precisely defining the concept of smart energy (SE), we contribute to the discussion on the challenges related to “territorially oriented energy management” policy. On the basis of our research, we take the position that cooperation between border communes in terms of energy challenges is becoming extremely relevant. This study draws on quantitative and qualitative data. We undertook a comparative study of energy production and consumption in Poland and Germany, thus presenting a regional and local perspective. The main research outcomes are conclusions resulting from the comparison of different strategies in the energy areas of Poland and Germany, a definition of the term SE, and solutions concerning interregional, and especially cross-border, cooperation. One of the key conclusions of this study is that in practice regional asymmetry in the Polish-German borderland may stimulate wider cross-border cooperation based on the SE concept.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
By analyzing the regional disparities in energy production and consumption between Poland and Germany, and precisely defining the concept of smart energy (SE), we contribute to the discussion on the challenges related to “territorially oriented energy management” policy. On the basis of our research, we take the position that cooperation between border communes in terms of energy challenges is becoming extremely relevant. This study draws on quantitative and qualitative data. We undertook a comparative study of energy production and consumption in Poland and Germany, thus presenting a regional and local perspective. The main research outcomes are conclusions resulting from the comparison of different strategies in the energy areas of Poland and Germany, a definition of the term SE, and solutions concerning interregional, and especially cross-border, cooperation. One of the key conclusions of this study is that in practice regional asymmetry in the Polish-German borderland may stimulate wider cross-border cooperation based on the SE concept.
5.
Chirkowska-Smolak, Teresa; Górecki, Tomasz; Klakus, Mateusz; Metzger, Weronika; Szargan, Martyna
The factorial validity of the Maslach Burnout Inventory: Student Survey (MBI-SS) in Poland Journal Article
In: Polish Psychological Bulletin, vol. 54, no. 3, pp. 207-216, 2023.
@article{Chirkowska2023,
title = {The factorial validity of the Maslach Burnout Inventory: Student Survey (MBI-SS) in Poland },
author = {Teresa Chirkowska-Smolak and Tomasz Górecki and Mateusz Klakus and Weronika Metzger and Martyna Szargan},
doi = {10.24425/ppb.2023.148053},
year = {2023},
date = {2023-12-31},
urldate = {2023-12-31},
journal = {Polish Psychological Bulletin},
volume = {54},
number = {3},
pages = {207-216},
abstract = {This paper explores the psychometric properties of the Polish adaptation of the MBI-SS questionnaire, which measures academic burnout in students. The factorial structure of the tool was examined, and its validity was evaluated using a sample from Polish state universities (N=935). Findings suggest that the MBI-SS possesses strong psychometric parameters. Data supports the tool's 3-factor structure and internal consistency. However, there are reservations regarding one item (MBI_6). Consequently, the 15-item version of the MBI-SS is recommended as a suitable tool for assessing burnout in Polish students},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper explores the psychometric properties of the Polish adaptation of the MBI-SS questionnaire, which measures academic burnout in students. The factorial structure of the tool was examined, and its validity was evaluated using a sample from Polish state universities (N=935). Findings suggest that the MBI-SS possesses strong psychometric parameters. Data supports the tool's 3-factor structure and internal consistency. However, there are reservations regarding one item (MBI_6). Consequently, the 15-item version of the MBI-SS is recommended as a suitable tool for assessing burnout in Polish students
4.
Łaźniewska, Ewa; Górecki, Tomasz; Kurowska-Pysz, Joanna
The impact of the Covid-19 pandemic on the resilience of the labour market in the Polish-German borderland Journal Article
In: Eastern Journal of European Studies, vol. 14, iss. 2, pp. 178-199, 2023.
@article{Lazniewska2023b,
title = {The impact of the Covid-19 pandemic on the resilience of the labour market in the Polish-German borderland},
author = {Ewa Łaźniewska and Tomasz Górecki and Joanna Kurowska-Pysz},
url = {https://ejes.uaic.ro/articles/EJES2023_1402_LAZ.pdf},
doi = {10.47743/ejes-2023-0208},
year = {2023},
date = {2023-12-01},
urldate = {2023-12-01},
journal = {Eastern Journal of European Studies},
volume = {14},
issue = {2},
pages = {178-199},
abstract = {The COVID-19 pandemic quite strongly disrupted the existing trends in the evolution of employment in the Polish-German borderland. The objective of the article is to analyse how resilient Polish and German municipalities are to the COVID-19 pandemic. Border regions, due to their specificities, are characterized by weaker resilience to negative external factors, which meant that the COVID-19 pandemic had a detrimental effect and caused an upturn in unemployment. The empirical part of the study concerns the Polish- German borderland as a clear example of the described tendencies. The research method used in the study is the counterfactual before-after comparison method, which fills the gap in the literature regarding research of cross border regions and develops the existing approaches. This approach is novel in studying the hypothetical rate of unemployment. The results of the survey indicate that the unemployment rate in the Polish-German borderland area has risen (after and due to the pandemic There are significant differences in how the pandemic has affected the labor market. Utilizing the counterfactual approach, this difference can be described as the impact of the COVID- 19 pandemic on the labor market.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The COVID-19 pandemic quite strongly disrupted the existing trends in the evolution of employment in the Polish-German borderland. The objective of the article is to analyse how resilient Polish and German municipalities are to the COVID-19 pandemic. Border regions, due to their specificities, are characterized by weaker resilience to negative external factors, which meant that the COVID-19 pandemic had a detrimental effect and caused an upturn in unemployment. The empirical part of the study concerns the Polish- German borderland as a clear example of the described tendencies. The research method used in the study is the counterfactual before-after comparison method, which fills the gap in the literature regarding research of cross border regions and develops the existing approaches. This approach is novel in studying the hypothetical rate of unemployment. The results of the survey indicate that the unemployment rate in the Polish-German borderland area has risen (after and due to the pandemic There are significant differences in how the pandemic has affected the labor market. Utilizing the counterfactual approach, this difference can be described as the impact of the COVID- 19 pandemic on the labor market.
2020
3.
Górecki, Tomasz; Horváth, Lajos; Kokoszka, Piotr
Tests of normality of functional data Journal Article
In: International Statistical Review , vol. 88, no. 3, pp. 677-697, 2020.
@article{Gorecki2020,
title = {Tests of normality of functional data},
author = {Tomasz Górecki and Lajos Horváth and Piotr Kokoszka},
doi = {10.1111/insr.12362},
year = {2020},
date = {2020-02-17},
urldate = {2020-02-17},
journal = {International Statistical Review },
volume = {88},
number = {3},
pages = {677-697},
abstract = {The paper is concerned with testing normality in samples of curves and error curves estimated from functional regression models. We propose a general paradigm based on the application of multivariate normality tests to vectors of functional principal components scores. We examine finite sample performance of a number of such tests and select the best performing tests. We apply them to several extensively used functional data sets and determine which can be treated as normal, possibly after a suitable transformation. We also offer practical guidance on software implementations of all tests we study and develop large sample justification for tests based on sample skewness and kurtosis of functional principal component scores.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The paper is concerned with testing normality in samples of curves and error curves estimated from functional regression models. We propose a general paradigm based on the application of multivariate normality tests to vectors of functional principal components scores. We examine finite sample performance of a number of such tests and select the best performing tests. We apply them to several extensively used functional data sets and determine which can be treated as normal, possibly after a suitable transformation. We also offer practical guidance on software implementations of all tests we study and develop large sample justification for tests based on sample skewness and kurtosis of functional principal component scores.
2018
2.
Górecki, Tomasz; Horváth, Lajos; Kokoszka, Piotr
Change point detection in heteroscedastic time series Journal Article
In: Econometrics and Statistics, vol. 7, pp. 63-88, 2018.
@article{Gorecki2018,
title = {Change point detection in heteroscedastic time series},
author = {Tomasz Górecki and Lajos Horváth and Piotr Kokoszka},
doi = {10.1016/j.ecosta.2017.07.005},
year = {2018},
date = {2018-01-01},
urldate = {2018-01-01},
journal = {Econometrics and Statistics},
volume = {7},
pages = {63-88},
abstract = {Many time series exhibit changes both in level and in variability. Generally, it is more important to detect a change in the level, and changing or smoothly evolving variability can confound existing tests. A framework for testing for shifts in the level of a series which accommodates the possibility of changing variability is developed. The resulting tests are robust both to heteroskedasticity and serial dependence. They rely on a new functional central limit theorem for dependent random variables whose variance can change or trend in a substantial way. This new result is of independent interest as it can be applied in many inferential contexts applicable to time series. Its application to change point tests relies on a new approach which utilizes Karhunen–Loéve expansions of the limit Gaussian processes. After presenting the theory in the most commonly encountered setting of the detection of a change point in the mean, it is shown how it can be extended to linear and nonlinear regression. Finite sample performance is examined by means of a simulation study and an application to yields on US treasury bonds.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Many time series exhibit changes both in level and in variability. Generally, it is more important to detect a change in the level, and changing or smoothly evolving variability can confound existing tests. A framework for testing for shifts in the level of a series which accommodates the possibility of changing variability is developed. The resulting tests are robust both to heteroskedasticity and serial dependence. They rely on a new functional central limit theorem for dependent random variables whose variance can change or trend in a substantial way. This new result is of independent interest as it can be applied in many inferential contexts applicable to time series. Its application to change point tests relies on a new approach which utilizes Karhunen–Loéve expansions of the limit Gaussian processes. After presenting the theory in the most commonly encountered setting of the detection of a change point in the mean, it is shown how it can be extended to linear and nonlinear regression. Finite sample performance is examined by means of a simulation study and an application to yields on US treasury bonds.
1.
Górecki, Tomasz; Hörmann, Siegfried; Horváth, Lajos; Kokoszka, Piotr
Testing normality of functional time series Journal Article
In: Journal of Time Series Analysis, vol. 39, iss. 4, pp. 471-487, 2018.
@article{Gorecki2018b,
title = {Testing normality of functional time series},
author = {Tomasz Górecki and Siegfried Hörmann and Lajos Horváth and Piotr Kokoszka},
doi = {10.1111/jtsa.12281},
year = {2018},
date = {2018-01-01},
urldate = {2018-01-01},
journal = {Journal of Time Series Analysis},
volume = {39},
issue = {4},
pages = {471-487},
abstract = {We develop tests of normality for time series of functions. The tests are related to the commonly used Jarque–Bera test. The assumption of normality has played an important role in many methodological and theoretical developments in the field of functional data analysis. Yet, no inferential procedures to verify it have been proposed so far, even for i.i.d. functions. We propose several approaches which handle two paramount challenges: (i) the unknown temporal dependence structure and (ii) the estimation of the optimal finite-dimensional projection space. We evaluate the tests via simulations and establish their large sample validity under general conditions. We obtain useful insights by applying them to pollution and intraday price curves. While the pollution curves can be treated as normal, the normality of high-frequency price curves is rejected.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We develop tests of normality for time series of functions. The tests are related to the commonly used Jarque–Bera test. The assumption of normality has played an important role in many methodological and theoretical developments in the field of functional data analysis. Yet, no inferential procedures to verify it have been proposed so far, even for i.i.d. functions. We propose several approaches which handle two paramount challenges: (i) the unknown temporal dependence structure and (ii) the estimation of the optimal finite-dimensional projection space. We evaluate the tests via simulations and establish their large sample validity under general conditions. We obtain useful insights by applying them to pollution and intraday price curves. While the pollution curves can be treated as normal, the normality of high-frequency price curves is rejected.