Przeglądaj wg Słowo kluczowe "sieci neuronowe"

Teraz wyświetlane 1 - 3 z 3
  • Miniatura
    Pozycja
    Employing neural networks to predict the number of incidents on specific types of Polish roads
    (Oficyna Wydawnicza AFM, 2024) Gorzelańczyk, Piotr; Drzewiecki, Janusz
    The article’s goal is to predict how many accidents will occur on different types of roads in Poland. This was accomplished by the analysis of annual data on the number of traffic accidents in Poland by type of road. A prediction for the years 2022–2040 was developed using police statistics. The frequency of accidents in Poland was anticipated using a few neural network models. The findings indicate that we can still expect a stabilization of the number of road accidents. This is impacted by the rise in traffic on Polish roads and the construction of new highways. The number of learning, test, and validation samples chosen at random has an impact on the outcomes.
  • Miniatura
    Pozycja
    Folia Oeconomica Cracoviensia, Vol. LII
    (Krakowska Akademia im. Andrzeja Frycza Modrzewskiego, Polska Akademia Nauk - Oddział w Krakowie - Komisja Nauk Ekonomicznych i Statystyki, 2011) Baster, Paweł; Pocztowska, Katarzyna; Malczyk, Krzysztof; Makieła, Kamil; Osiewalski, Jacek; Osiewalski, Krzysztof; Prusak, Anna; Stefanów, Piotr; Czarnecki, Lech; Iwasiewicz, Andrzej
  • Miniatura
    Pozycja
    Sieci neuronowe i polichotomiczne modele zmiennych jakościowych w analizie ryzyka kredytowego
    (Oficyna Wydawnicza AFM, 2011) Baster, Paweł; Pocztowska, Katarzyna
    Management of credit risk, one of the main bank activities, is currently a very important issue. This paper contains comparison of two instruments used in prediction of probability that consumer fails to fully repay a loan in agreed time: artificial neural networks and models for polychotomous ordered data. For the empirical research each client has been assigned to one of four categories reflecting his/her delay in payments. Estimation and validation of methods was performed on a 3000-item sample containing information about each loan agreement and repayment history originating from one of Polish banks, covering years 2000-2001. The dataset was repeatedly divided into train and validation sets. Multi-layer architecture of artificial neural network with logistic activation function was proposed. Ordered logit and probit models were estimated within maximum likelihood framework. Several alternative specifications were proposed differing in independent variable set (including their products and squares). Bank income was chosen as the main criterion of fitness. Problem of optimal decision and defining appropriate loss function was formulated on the basis of statistical decision theory. Furthermore, properties of estimated models related to inference about probability of repayment and credit risk factors were presented.