This study intends to provide a new insight into the concentration and dominance indices as the concerns grow about the increasing concentration in the markets around the world. Most of the studies attempting to measure concentration or dominance in a market employ the popular concentration/dominance indices like Herfindahl–Hirschmann, Hannah–Kay, Rosenbluth–Hall–Tidemann and Concentration ratio. On the other hand, measures of qualitative variation are closely related to entropy, diversity and concentration/dominance measures. In this study, two normalized dominance measures that can be derived from the work of Wilcox on qualitative variation are proposed. The limiting distributions of these normalized dominance measures are formulated. By some simulations, asymptotic behaviors of these indices are analyzed under some assumptions about the market structure. In the end, by an application on the Turkish car sales in 2019, it is determined that the values of dominance indices vary in a considerably large range. Thus one of the dominance indices is determined to have the advantage of having less error in estimation, less sensitivity to smaller market shares, and less sampling variability.
26 December 2022 in Dokuz Eylül Üniversitesi Sosyal Bilimler Enstitüsü Dergisi
Çalışma, NFT (Non-FungibleToken) kavramı ile bu kavramın üzerine kurulu olduğu teknoloji ve kullanım alanlarını araştırmaktadır. Bu araştırmada ilk olarak NFT’nin dayandığı blokzincir teknolojisi incelenmekte ve çalışma biçimi açıklanmaktadır. Ardından bu teknolojinin özellikleri ve popüler kripto paralar dışındaki kullanım alanları irdelenmiştir. Daha sonra NFT’yi bir standart olarak tanımlayan Ethereum platformu araştırılmış ve yine bu platform üzerinde tanımlanan akıllı sözleşmeler açıklanmıştır. Ayrıca NFT kategorileri incelenmiş, dijital sanat alanı ve diğer kategorilerde önerilen uygulamalar için bir literatür taraması yapılmıştır. Ek olarak, oluşturulan sözlük yardımıyla bu alanda kullanılan kavramların Türkçe karşılıklarına dair önerilerde bulunulmuştur. Başta sanat, koleksiyon ve oyun kategorisinde olmak üzere birçok alanda NFT uygulamaları mevcuttur. NFT’lerin özellikle tedarik zincirleri ve nesnelerin interneti alanlarında yaygın olarak kullanım alanı bulacağı öngörülmektedir. Bu çalışma, NFT kavramını, teknolojisini ve kullanım alanlarını bir bütün olarak ele almakta ve açıklamaktadır.
Asymptotic behavior of qualitative variation statistics, including entropy measures, can be modeled well by normal distributions. In this study, we test the normality of various qualitative variation measures in general. We find that almost all indices tend to normality as the sample size increases, and they are highly correlated. However, for all of these qualitative variation statistics, maximum uncertainty is a serious factor that prevents normality. Among these, we study the properties of two qualitative variation statistics; VarNC and StDev statistics in the case of maximum uncertainty, since these two statistics show lower sampling variability and utilize all sample information. We derive probability distribution functions of these statistics and prove that they are consistent. We also discuss the relationship between VarNC and the normalized form of Tsallis (α = 2) entropy in the case of maximum uncertainty.
Feature selection is a critical preprocessing step for mitigating the curse of dimensionality in machine learning. Existing methods present a difficult trade-off: filter methods are fast but often suboptimal as they evaluate features in isolation, while wrapper methods are powerful but computationally prohibitive due to their iterative nature. In this paper, we propose LAFS (Learnable Attention for Feature Selection), a novel, end-to-end differentiable framework that achieves the performance of wrapper methods at the speed of simpler models. LAFS employs a neural attention mechanism to learn a context-aware importance score for all features simultaneously in a single forward pass. To encourage the selection of a sparse and non-redundant feature subset, we introduce a novel hybrid loss function that combines the standard classification objective with an information-theoretic entropic regularizer on the attention weights. We validate our approach on real-world high-dimensional benchmark datasets. Our experiments demonstrate that LAFS successfully identifies complex feature interactions and handles multicollinearity. In general comparison, LAFS achieves very close and accurate results to state-of-the-art RFE-LGBM and embedded FSA methods. Our work establishes a new point on the accuracy-efficiency frontier, demonstrating that attention-based architectures provide a compatible solution to the feature selection problem.
Mean-variance portfolio optimization model has been shown to have serious drawbacks. The model assumes that assets returns are normally distributed that is not valid for most of the markets and portfolios. It also relies on asset’s covariance matrices for the calculation of portfolio’s risk that is open to estimation errors. Moreover, these optimization errors are maximized by the method that result in poor out-of-sample performances. In this study, we propose a new portfolio optimization method based on minimization of Tsallis entropy, which is valid for any underlying distribution. First, we show that the Tsallis entropy can be employed as a risk measure for portfolio analysis. Then we demonstrate the validity of the model by comparing its performance with those mean-variance and minimum-variance portfolios using BIST 30 data.
Graphical Lasso (Least absolute shrinkage and selection operator) has become a popular tool in the field of machine learning in recent years. Although it has been deployed mainly for feature selection in classification problems, it is also used for covariance matrix estimation. Mean-variance portfolio optimization relies on sample covariance matrix for the calculation of the portfolio’s risk, whereas it has been most hardly criticized. The aim of this study is to demonstrate the effect of the covariance matrix estimation by Graphical Lasso algorithm with varying L1 penalty factors. Mean-variance portfolio optimization using empirical and estimated covariance matrices are applied to BIST 30 index and the results are compared.