Hisse senetleri pazarını yenme modelleri ve İ.M.K.B. üzerinde yapay zeka uygulamaları
Başlık çevirisi mevcut değil.
- Tez No: 71806
- Danışmanlar: PROF. DR. NİYAZİ BERK
- Tez Türü: Doktora
- Konular: İşletme, Business Administration
- Anahtar Kelimeler: Analiz, Borsa, Hisse senetleri, Teknik analiz, Temel analiz, Yapay zeka, İMKB, Analysis, Stock exchange, Stocks, Technical analysis, Basic analysis, Artificial intelligence, İstanbul Stock Exchange
- Yıl: 1998
- Dil: Türkçe
- Üniversite: Marmara Üniversitesi
- Enstitü: Bankacılık ve Sigortacılık Enstitüsü
- Ana Bilim Dalı: Bankacılık Ana Bilim Dalı
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: Belirtilmemiş.
Özet
Özet yok.
Özet (Çeviri)
Plenty of technical and fundamental analysis indicators have been evolved so far. Unfortunately, these indicators often cause false decisions For instance, while a moving average indicator may give a sell signal a momentum indicator may give a buy signal. Hence, skill of technical analysts is very vital and may change decisions people. Furthermore, firms are supposed to be powerful in terms of financial statements which give access to good Performa's in the stock markets. Although there are lots of suspicions of technical and fundamental analysis, of course, these are both useful. But these need knowledgements and experiences. Furthermore, investors needs more objective measures. Stock exchanges are effected by many factors. So investors have to investigate these factors simultaneously. Classification analysis may be used to obtain this object successfully. Main analysis techniques to classification below are; 01 regression analysis; this is a kind of simplest classification analysis. The parameters of all linear probability model might be estimated by the least squared method. Multiple discriminant analysis models may be linear or quadratic. There are lots of algorithms that have been evolved. But the matrix method is the most ideal method. Multiple quadratic discriminant: Although that model is much more complex it could not be more successful than linear disciriminant model. Accumulated probability models are similar to each other except the excess values. And are more successful than regression and discriminant analysis. Nevertheless, all of these techniques are very common in practice, they could not give high accuracy for stock evaluation. Because they can not adjust their errors. Contrarily, human can have experiences from errors. Human brain is consisted of one a thousand billion cells named noron. Norons are very simple. A noron cells consists of three parts. These are nucleus, synapses, cell wall. Human brain uses database that contains historical expriences and often decide wrongly. Human brain inpects factors that effect their decision and varies counts and weights of those factors. So ifs decisions getting nearly perfect performance more and more.
Benzer Tezler
- Portföy yönetimi öi markowitz yaklaşımının değerlendirilmesi
Portfolio management and markowitz theory
DOĞAN YAVUZ
- A Study on the price behaviour of initial public afferings
İlk kez halka arz edilen hisse senetlerinin fiyat davranışları üzerine bir çalışma
MEHMET HAKAN ERDEM
