Now showing items 1-20 of 22

    • A Proposal for Using Selected Tree-Based Models to Identify Operative Risk Subgroups among Patients Undergoing Coronary Artery Bypass Grafting 

      Misztal, Małgorzata ORCID (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      Classification and regression trees are very popular and attractive types of classifiers, widely used to solve decision-making problems in different fields of science. The study was conducted to identify preoperative ...
    • About Phase Transitions in Kendall’s Shape Space 

      Kosiorowski, Daniel (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In our article, we discuss the choice of space shape, appropriate for describing an economic process and analyze usefulness of the metrics proposed by I. L. Drуden and K. V. Merida (1998). We introduce and interpret the ...
    • Analysis of Linear Regression Model at Divided System Matrix 

      Budka, Anna; Wagner, Wiesław (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In this study problems connected with the detection of influential observations are investigated in the linear regression model using the least squares estimation of structural parameters. This issue has been presented ...
    • Comparative Assessment of Some Selected Methods of Determining the Number of Clusters in a Data Set 

      Korzeniewski, Jerzy ORCID (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      This paper is an attempt to compare the performance of an algorithm for determining the number of clusters in a data set proposed by the author with other methods of determining the number of clusters. The idea of the ...
    • Discrimination of Symbolic Objects 

      Dudek, Andrzej (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      Symbolic Data Analysis is an extension of multivariate analysis dealing with data represented in an extended form. Each cell in symbolic data table (symbolic variable) can contain data in form of single quantitative ...
    • Estimation of Bias and Variance of Sample Median by Jacknife and Bootstrap Methods 

      Pruska, Krystyna (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In the paper the estimation of sample median bias and variance by jackknife and bootstrap methods are considered. Monte Carlo analysis of properties of estimators is presented (mean of bias and mean of variance for some ...
    • Feature Selection and Multiple Model Approach in Discriminant Analysis 

      Gatnar, Eugeniusz (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      Significant improvement of model stability and prediction accuracy in classification and regression can be obtained by using the multiple model approach. In classification multiple models are built on the basis of training ...
    • GARCH Models of Time Series on DAM 

      Ganczarek, Alicja (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In this paper an analysis of the time series on the Day Ahead Market (DAM) of the Polish Power Exchange is presented. In this analysis Generalized Autoregressive Conditional Heteroscedasticity (GARCH) models are used to ...
    • Introduction 

      Domański, Czesław; Pekasiewicz, Dorota (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
    • Jackknife Forecasts of Time Series 

      Wywiał, Janusz; Żądło, Tomasz (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In the paper we present the examples of forecasts of time series with seasonal fluctuations. Based on the jackknife method we estimate variances of seasonal factors and the MSE of prediction. Jackknife method has been ...
    • Joint Monitoring of the Mean, Dispersion and Asymmetry in Process Control 

      Kończak, Grzegorz (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In this paper the quadratic form of sample mean, sample variance and sample asymmetry is considered. This quadratic form can be used for testing hypothesis on the expected value, variance and asymmetry of observed variable. ...
    • Kriging - A Method of Statistical Interpolation of Spatial Data 

      Kowalik, Jan (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In analyses of spatial phenomena it can sometimes be impossible or very expensive to obtain the value (realization) of a studied phenomenon at all its locations because of practical constraints. Then in order to estimate ...
    • Methods of Classification Error Decompositions and their Properties 

      Rozmus, Dorota (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      The idea of error decomposition originates in regression where squared loss function is applied. More recently, several authors have proposed corresponding decompositions for classification problem, where 0-1 loss is ...
    • Multiple Additive Regression Trees (MART) and their Application 

      Trzęsiok, Joanna (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      Multiple additive regression trees MART is a methodology for trying to solve prediction problems in regression and classification. It’s one of the boosting methods. It was introduced by J. H. Friedman (1999a). Besides ...
    • Multiple Endpoints 

      Parys, Dariusz (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In ANOVA we are mainly based on inter-treatment comparisons. Another common problems arising in biometric studies (especially in biomedical studies) is that of comparing two groups of patients (treatment and a control ...
    • Multivalued Markov Processes 

      Trzpiot, Grażyna (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      Multivalued random variables and stochastic processes can be use in integral geometry, mathematical economics or stochastic optimization. Using the methods of selection operators we can give the selection characterization ...
    • On some Inferential Procedures for Receiver Operating Characteristic Curves 

      Rossa, Agnieszka ORCID (Wydawnictwo Uniwersytetu Łódzkiego, 2006)
      In the paper two significance tests for receiver operating characteristic curves (ROC) are proposed. Both tests use an asymptotic χ² distribution of the test statistics.
    • On the Application of Classification Trees to Analyze Customer Loyalty and Satisfaction 

      Michalak, Jarosław (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      It became more important to recognize customer expectations and to oiler product properties, which convince customer to buy the analyzed product. In customer loyalty and satisfaction analysis methods, classification trees ...
    • Sequential Tests with Power Equal to 1 for Chosen Location Parameters 

      Pekasiewicz, Dorota (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      We often verify hypotheses about random variable distribution parameters, when the variable distribution is unknown. In these cases we apply nonparametric tests, in particular nonparametric sequential tests. This paper ...
    • Some Nonparametric Estimators of Regression Function 

      Baszczyńska, Aleksandra ORCID (Wydawnictwo Uniwersytetu Łódzkiego, 2007)
      In the paper some nonparametric estimators of regression function are studied: Nadaraya-Watson estimator and k-nearest neighbour one. Properties of these estimators and possibilities of using them in practice are taken ...