Wyniki wyszukiwania - Biblioteka Nauki

1

Pattern recovery and signal denoising by SLOPE when the design matrix is orthogonal

100%

Skalski T. , Kołodziejek B. , Graczyk P. , Wilczyński M.

Probability and Mathematical Statistics

|

2022

|

tom Vol. 42, Fasc. 2

283--302

EN

Sorted ℓ1 Penalized Estimator (SLOPE) is a relatively new convex regularization method for fitting high-dimensional regression models. SLOPE allows the reduction of the model dimension by shrinking some estimates of the regression coefficients completely to zero or by equating the absolute values of some nonzero estimates of these coefficients. This allows one to identify situations where some of true regression coefficients are equal. In this article we will introduce the SLOPE pattern, i.e., the set of relations between the true regression coefficients, which can be identified by SLOPE. We will also present new results on the strong consistency of SLOPE estimators and on the strong consistency of pattern recovery by SLOPE when the design matrix is orthogonal and illustrate advantages of the SLOPE clustering in the context of high frequency signal denoising.

2

An accurate model of LED luminaire using measurement results for estimation of electrical parameters based on the multivariable regression method

100%

Sikora R. , Markiewicz P. , Rózga P.

|

2022

|

tom Vol. 29, nr 2

251--269

EN

Light sources and luminaires made in the LED technology are nowadays widely used in industry and at home. The use of these devices affects the operation of the power grid and energy efficiency. To estimate this impact, it is important to know the electrical parameters of light sources and luminaires, especially with the possibility of dimming. The article presents the results of measurements of electrical parameters as well as luminous flux of dimmable LED luminaires as a function of dimming and RMS supply voltage. On the basis of the performed measurements, a model of LED luminaire was developed for prediction of electrical parameters at set dimming values and RMS values of the supply voltage. The developed model of LED luminaire has 2 inputs and 26 outputs. This model is made based on 26 single models of electrical parameters, whose input signals are supply and control voltages. The linear regression method was used to develop the models. An example of the application of the developed model for the prediction of electrical parameters simulating the operation of an LED luminaire in an environment most similar to real working conditions is also presented.

3

Untypical Observations in Linear Regression

100%

Kobylińska M.

|

2011

|

tom 6

|

nr 2

239-249

EN

In the analysed set of socioeconomic phenomena and processes results differing from the others may occur. Revealing such untypical observations is an important research issue as they may distort the statistical analysis of the investigated phenomenon. The paper discusses the types of untypical observations in two-dimensional sample. The method for detecting untypical observations in linear regression based on the measures of observation depth in the sample was proposed that was illustrated on the base of a numeric example.

4

Fuzzy regression and its application for data analysis

100%

Walaszek-Babiszewska A.

Management

|

2002

|

tom Vol. 6, no 1

121-125

EN

In this paper we present fuzzy linear regression for estimating of parameters. The regression method has been used to creating the fuzzy description of some economic data.

PL

W referacie przedstawiono podstawy regresji liniowej sformułowane w kategoriach zbiorów rozmytych. Metoda rozmytej regresji liniowej została zastosowana do określenia modelu liniowego pewnych danych ekonomicznych.

5

Dependent noise for stochastic algorithms

100%

Doukhan P. , Brandierie O.

Probability and Mathematical Statistics

|

2004

|

tom Vol. 24, Fasc. 2

381--399

EN

We introduce different ways of being dependent for the input noise of stochastic algorithms. We are aimed to prove that such innovations allow to use the ODE (ordinary differential equation) method. Illustrations to the linear regression frame and to the law of large numbers for triangular arrays of weighted dependent random variables are also given.

6

EA-MOSGWA : a tool for identifying associated SNPs in Genome Wide Association Studies

100%

Gola A. , Bogdan M. , Frommlet F.

Theoretical and Applied Informatics

|

2013

|

tom Vol. 25, No. 3-4

251--262

EN

This paper presents the current stage of the development of EA-MOSGWA - a tool for identifying causal genes in Genome Wide Association Studies (GWAS). The main goal of GWAS is to identify chromosomal regions which are associated with a particular disease (e.g. diabetes, cancer) or with some quantitative trait (e.g height or blood pressure). To this end hundreds of thousands of Single Nucleotide Polymorphisms (SNP) are genotyped. One is then interested to identify as many SNPs as possible which are associated with the trait in question, while at the same lime minimizing the number of false detections. The software package MOSGWA allows to detect SNPs via variable selection using the criterion mBIC2, a modified version of the Schwarz Bayesian Information Criterion. MOSGWA tries to minimize mBIC2 using some stepwise selection methods, whereas EA-MOSGWA applies some advanced evolutionary algorithms to achieve the same goal. We present results from an extensive simulation study where we compare the performance of EA-MOSGWA when using different parameter settings. We also consider using a clustering procedure to relax the multiple testing correction in mBlC2. Finally we compare results from EA-MOSGWA with the original stepwise search from MOSGWA, and show that the newly proposed algorithm has good properties in terms of minimizing the mBIC2 criterion, as well as in minimizing the misclassification rate of detected SNPs.

PL

W artykule przedstawiony jest aktualny stan rozwoju programu EA-MOSGWA - narzędzia służącego do identyfikacji przyczynowych genów w badaniach asocjacyjnych całego genomu (ang. Genome Wide Association Studies, GWAS). Głównym celem tych badań jest określenie tych rejonów chromosomu, które są związane z występowaniem chorób genetycznych (np. cukrzyca, rak) lub wpływają na daną cechę (np. wysokość lub ciśnienie krwi). Sprowadzają się one do przebadania wielu tysięcy polimorfizmów pojedynczego nukleotydu (ang. Single Nucleotide Polymorphisme SNP) i powiązaniu ich (pojedynczych lub grupy SNPów) z przypadkami klinicznymi oraz możliwymi do zmierzenia cechami. Kluczową kwestią jest zidentyfikowanie jak największej liczby przyczynowych SNPów przy jednoczesnej minimalizacji fałszywych odkryć. Program MOSGWA umożliwia detekcje SNPów poprzez wybór zmiennych z użyciem kryterium mBIC2 - zmodyfikowanej wersji Bayesowskiego kryterium informacyjnego Schwarza. MOSGWA stara się zminimalizować mBIC2 przy pomocy metody selekcji Stepwise, podczas gdy EA-MOSGWA wykorzystuje w tym cclu zmodyfikowaną wersję algorytmu ewolucyjnego. W artykule prezentujemy wyniki szeroko zakrojonych badań symulacyjnych, w których możemy porównać wydajność EA-MOSGWA przy użyciu różnych ustawień parametrów. Również bierzemy pod uwagę klasteryzację SNPów, aby złagodzić korekcje wielokrotnego testowania w metodzie mBIC2. Przedstawiamy także porównanie wyników otrzymanych przez EA-MOSGWA z wynikami metody Stepsiwe używanej w programie MOSGWA, aby pokazać że proponowana metoda ma dobre właściwości minimalizacji kryterium mBIC2 oraz minimalizacji wskaźnika fałszywych detekcji.

7

An Ancillary Paradox in Testing

100%

He K.

Probability and Mathematical Statistics

|

1999

|

tom Vol. 19, Fasc. 1

15--21

EN

In multiple linear regression with normally distributed errors, it is shown that a test procedure for a hypothesis about the intercept which is α-admissible when the design matrix is fixed is inadmissible when the design matrix is an ancillary statistic. The result of this paper is a complementary one to Brown's paper [2].

8

EA-MOSGWA : a tool for identifying associated SNPs in Genome Wide Association Studies

100%

Gola A. , Bogdan M. , Frommlet F. , Gola A. , Bogdan M. , Frommlet F.

|

nr 3-4

9

Pseudo Bayesian and Linear Regresssion Global Thresholding

100%

Aboura K.

|

tom Vol. 56, No. 1

63-72

EN

Classification is an important task in image analysis. Simply recognizing an object in an image can be a daunting step for a computer algorithm. The methodologies are often simple but rely heavily on the thresholding of the image. The operation of turning a color or gray-scale image into a black and white image is a determining step in the effectiveness of a solution. Thresholding methods perform differently in various problems where they are often used locally. Global thresholding is a difficult task in most problems. We highlight a pseudo Bayesian and a linear regression global thresholding methods that performed well in an engineering problem. The same approaches can be used in biomedical applications where the environment is better controlled.

10

Critical points of hypoeutectoid steel - prediction of the pearlite dissolution finish temperature Ac1f

100%

Pawłowski B.

Journal of Achievements in Materials and Manufacturing Engineering

|

2011

|

tom Vol. 49, nr 2

331--337

EN

Purpose: of this work is to present possibility of calculation of pearlite dissolution finish temperature during heating of hypoeutoctoid steels. Design/methodology/approach: The presented multiple linear regression equations for calculating the Ac1f temperature are based on experimental data set containing chemical composition and values of critical temperatures obtained by use of the dilatometric technique at the own laboratory only. Findings: The elaborated multiple linear regression equations for calculating the critical temperatures are an alternative to dilatometric examinations to obtain data necessary for proper heat treatment conditions planning. Research limitations/implications: All presented equations for calculating pearlite dissolution finish temperature are limited by range of mass concentrations of elements which is a consequence of limited data set used for elaboration of these equations. The obtained relationships do not concern other factors influencing Aclf temperature such as heating rate, grain size and interlamellar spacing of pearlite. Practical implications: Broadening the knowledge on the chemical composition influence on the critical temperatures, which will help in designing heat treatment conditions, especially of the Dual Phase steels. Originality/value: An attempt was made to find out a multiple linear regression formula between chemical composition and the pearlite dissolution finish temperature of hypoeutectoid steels.

11

Gas entrainment rate and flow characterization in downcomer of a Jameson cell

100%

Tasdemir T. , Tasdemir A. , Oteyaka B.

|

tom Vol. 47

61-78

EN

The Jameson cell which is a new type of gas-liquid contacting device and can be considered as a type of plunging jet column, has been in use worldwide for the separation of fine minerals, coal particles and wastewater treatment etc. Flow characteristics in the downcomer of a Jameson cell are very important since the hydrodynamics of the cell is largely depends on the flow conditions. The hydrodynamics influences flow regimes in the downcomer and hence the gas holdup and bubble diameter are strongly affected by flow conditions. Depending on the air entrainment rate entered to the system, different flow regimes are observed in the downcomer. Bubbly flow which is observed at less air quantities is desired instead of churn-turbulent flow where the gas entrainment rate increase. In this research, the effect of operating conditions including nozzle diameter, downcomer diameter, jet velocity and jet length on gas entrainment rate, Qg , was evaluated experimentally for an air-water system for the bubbly and churn-turbulent flow. Between these factors, downcomer diameter was found to have very little effect on gas entrainment rate while increasing values of other factors had an increasing effect on it. The results were evaluated by forward stepwise linear regression (MLR) and a piecewise regression with Quasi-Newton estimation of breakpoint (PLR) to estimate the flow conditions and gas entrainment rates. The model by PLR was useful to understand the boundary of the flow characteristics since the two equations were valid in a certain air entrainment ranges, i.e. different flow conditions. The model developed was successful to determine the transition region from bubbly flow to churn-turbulent flow. Experimental data were in good agreement with theoretically predicted value.

12

ANALYSIS OF CALENDAR EFFECTS IN MARKETS OF PRECIOUS METALS

100%

Krawiec M. , Górska A.

Metody Ilościowe w Badaniach Ekonomicznych

|

2014

|

tom 15

|

nr 2

392-402

EN

Calendar effects are anomalies in the behavior of asset prices that may disprove the efficient market hypothesis. The well recognized are: day-of-the-week effect, month-of-the-year effect, holidays effect and turn-of-the-month effect. These anomalies are observed in many financial markets, most often on stock exchanges, thus studies on calendar effects usually focus on stock markets. However, the aim of the paper is searching for the anomalies in precious metals markets (the empirical data covers London daily spot prices from 2008 through 2013). This is the continuation of authors’ prior research aimed at testing weak market efficiency hypothesis for precious metals markets.

13

The Interaction of Chlorophyll-a and Total Suspended Matter along the Western Semarang Bay, Indonesia, Based on Measurement and Retrieval of Sentinel 3

100%

Maslukah L. , Ismunarti D. H. , Widada S. , Sandi N. F. , Prayitno H. B.

Journal of Ecological Engineering

|

2022

|

tom Vol. 23, nr 10

191--201

EN

The Kendal Regency area is one of the areas on the northern coast of Central Java that has been experiencing rapid industrial development. The high human activity in this area will impact the quality of water in these surrounding areas and affect the fertility of the waters. The concentrations of chlorophyll-a (Chl-a) and total suspended matter (TSM) are major water quality parameters that can be retrieved using remotely sensed data. The retrieval satellite of the 3 OLCI chosen in this study has a 300 m spatial resolution. This study aimed to see the distribution and effect of total suspended matter (TSM) on chlorophyll-a based on measurement and retrieval of Sentinel 3 imagery using the linear regression method. The results show the chlorophyll-a distribution and the value from retrieval satellite are higher and occur over larger surface area compared to chlorophyll-a measurements. The linear regression model of chlorophyll-a by retrieval satellite imagery and measurement is y = 0.65x + 4.65 with R2 = 0.54. The presence of high amounts of suspended solids in the waters causes disturbances in the reflectance values, which are recorded by the retrieval of satellite. The model regression chlorophyll-a with TSM accuracy from retrieval satellite results in the equation y = -0.0416x + 5.14 (R2 = 0.45, p = 0.05, n = 13). The determination (R2) coefficient value is 0.445, which means that suspended solids have a 44.5% effect on chlorophyll-a and 55.5% is influenced by other factors and not examined in this study. The results show that TSM has an influence on the accuracy of chlorophyll-a and retrieval satellite recording can be disrupted if waters have high turbidity.

14

Generalized ordered linear regression with regularization

88%

Łęski J. , Henzel N.

|

2012

|

tom Vol. 60, nr 3

481-489

EN

Linear regression analysis has become a fundamental tool in experimental sciences. We propose a new method for parameter estimation in linear models. The 'Generalized Ordered Linear Regression with Regularization' (GOLRR) uses various loss functions (including the o-insensitive ones), ordered weighted averaging of the residuals, and regularization. The algorithm consists in solving a sequence of weighted quadratic minimization problems where the weights used for the next iteration depend not only on the values but also on the order of the model residuals obtained for the current iteration. Such regression problem may be transformed into the iterative reweighted least squares scenario. The conjugate gradient algorithm is used to minimize the proposed criterion function. Finally, numerical examples are given to demonstrate the validity of the method proposed.

15

Measuring the stray capacitance of solenoids with a transmitting and a receiving coil

88%

Mariscotti A.

|

tom Vol. 18, nr 1

47-56

EN

Air core solenoids, possibly single layer and with significant spacing between turns, are often used to ensure low stray capacitance, as they are used as part of many sensors and instruments. The problem of the correct estimation of the stray capacitance is relevant both during design and to validate measurement results; the expected value is so low to be influenced by any stray capacitance of the external measurement instrument. A simplified method is proposed that does not perturb the stray capacitance of the solenoid under test; the method is based on resonance with an external capacitor and on the use of a linear regression technique.

16

Research of Disturbance in Model of Linear Regression Estimated According to Criteria of Least Squares

88%

Wagner W. , Budka A. , Wiesław Wagner - Academy of Physical Education P. , Anna Budka - Academy of Agriculture P.

|

2003

|

tom 164

PL

W pracy przedstawiono wyniki analiz numerycznych związanych z estymacją stopnia zakłócenia współczynników występujących w równaniu prostej regresji.

17

Fractal Dimension as Robust Estimate of Low Carbon Steels Hardness

88%

Zając K. , Płatek K. , Wachel P. , Łatka L.

Advances in Science and Technology. Research Journal

|

2022

|

tom Vol. 16, no 5

335--344

EN

Application of computational methods in engineering and science constantly increases, which is also visible in sector of material science, often with promising results. In following paper, authors would like to propose fractal dimension, a mathematical method of quantifying self-similarity and complexity of spatial patterns, as robust method of hardness estimation of low carbon steels. A dataset of microstructure images and corresponding Vickers hardness measurements of S235JR steel under different delivery conditions was created. Then, three different computational methods for evaluation of materials hardness based on microstructure image were tested. In this paper those methods are called: (i) Otsu-based index, (ii) fractal dimension index and (iii) vision transformer index. The results were compared with method used in literature for similar problems. Comparison showed that fractal dimension performs better than other evaluated methods, in terms of median absolute error, which value was equal to 4.12 HV1, which is significantly lower than results achieved by Otsu-based index and vision transformer index, which were 4.49 HV1 and 5.07 HV1 respectively. Those results can be attributed to the relative robustness of fractal dimension index, when compared to other methods. Robust estimation is preferable, due to the high amount of noise in the dataset, which is a consequence of the nature of used material.

18

Topsoil texture maps based on calibration of soil electrical conductivity with soil datasets varying in size

88%

Stępień M. K. , Gozdowski D. , Bodecka E. , Groszyk J. , Rozbicki J. , Samborski S.

|

2017

|

tom 50

|

nr 2

EN

The purpose of the study was to verify the possibility of creation of reliable soil texture class (STC) maps of a topsoil based on a linear calibration of shallow (0-30cm) soil electrical conductivity (ECsh) with small datasets of soil samples with laboratory determined STC . ECsh values were calibrated against four datasets of soil samples. The smallest datasets (5-6 soil samples per field) were selected: 1) in an arbitrary way; or 2) based on the quartiles of ECsh values. A dataset of an intermediate size (11-17 points) and a full dataset of all ST data available (33-38 points) were also tested. For one field, the calibration with ECsh quartiles produced STC maps with greater agreement with field's status than the complete dataset of laboratory results. Although, the root mean square errors and mean absolute errors were greater for quartiles than for the other datasets. The ECsh values depended on the content of fine soil (<2 mm) fractions to a depth of 90 cm, so ECsh measurements are efficient in mapping the topsoil texture of fields with relatively uniform texture in subsoil. The datasets, which produced lower values of errors did not always permit to prepare more accurate STC maps.

19

Comparison of bayesian and other approaches to the estimation of fatigue crack growth rate from 2D textural features

88%

Mojzeš M. , Kukal J. , Lauschmann H.

Journal of Theoretical and Applied Mechanics

|

2017

|

tom Vol. 55 nr 4

1269--1278

EN

The fatigue crack growth rate can be explained using features of the surface of a structure. Among other methods, linear regression can be used to explain crack growth velocity. Nonlinear transformations of fracture surface texture features may be useful as explanatory variables. Nonetheless, the number of derived explanatory variables increases very quickly, and it is very important to select only few of the best performing ones and prevent overfitting at the same time. To perform selection of the explanatory variables, it is necessary to assess quality of the given sub-model. We use fractographic data to study performance of different information criteria and statistical tests as means of the sub-model quality measurement. Furthermore, to address overfitting, we provide recommendations based on a cross-validation analysis. Among other conclusions, we suggest the Bayesian Information Criterion, which favours sub-models fitting the data considerably well and does not lose the capability to generalize at the same time.

20

Pairwise versus Pointwise Ranking : A Case Study

75%

Melnikov V. , Gupta P. , Frick B. , Kaimann D. , Hüllermeier E.

|

tom Vol. 25

73--83

EN

Object ranking is one of the most relevant problems in the realm of preference learning and ranking. It is mostly tackled by means of two different techniques, often referred to as pairwise and pointwise ranking. In this paper, we present a case study in which we systematically compare two representatives of these techniques, a method based on the reduction of ranking to binary classification and so-called expected rank regression (ERR). Our experiments are meant to complement existing studies in this field, especially previous evaluations of ERR. And indeed, our results are not fully in agreement with previous findings and partly support different conclusions.