|
|
Afolabi, M.O. & Olude, O., 2007. Predicting Stock Prices Using a Hybrid Kohonen Self Oragnizing Map (SOM). Hawaii.
Abstract: A challenging and daunting task for financial investors is determining stock market timing—when to buy, sell and the future price of a stock. This challenge is due to the complexity of the stock market. New methods have emerged that increase the accuracy of stock prediction. Examples of these methods are Fuzzy logic, Neural Network and hybridized methods such as hybrid Kohonen Self Organizing Map (SOM), Adaptive Neuro–Fuzzy Inference System (ANFIS) etc. This paper presents a number of methods used to predict the stock price of the day. These methods are Backpropagation, Kohonen SOM, and a hybrid Kohonen SOM. The results show that the difference in error of the hybrid Kohonen SOM is significantly reduced compared to the other methods used. Hence, the results suggest that the hybrid Kohonen SOM is a better predictor compared to Kohonen SOM and Backpropagation.
Keywords: Clustering
|
|
|
|
Ailon, N., Chazelle, B., Comandur, S. & Liu, D.. Self-Improving Algorithms.
|
|
|
|
Dasgupta, S. & Hsu, D.. Hierarchical Sampling for Active Learning. Helsinki, Finland.
|
|
|
|
Deboeck, G. & Ultsch, A.. Picking Stocks with Emergent Self-organizing Value maps.
Abstract: Picking stocks that are suitable for portfolio management is a complex task. The most common criteria are the price earnings ratio, the price book ratio, price sales ratio, the price cash flow ratio, and market capitalization. Another approach called CAN SLIM relies on earnings growth (quarterly and annual earnings growth) of companies; the relative strength of the stock prices; the institutional sponsorship; the debt capital ratio, the shares outstanding, market capitalization, and the market direction. The main issue with the traditional approaches is the proper weighting of criteria to obtain a list of stocks that are suitable for portfolio management. This paper proposes an improved method for stock picking using the CAN SLIM system in conjunction with emergent self-organizing value maps to assemble a portfolio of stocks that outperforms a relevant benchmark. The neural network approach discussed in this paper finds structures in sets of stocks that fulfill the CAN SLIM criteria. These structures are visualized using UMatrix and used to construct portfolios. Portfolios constructed in this way perform better the more the CAN SLIM criteria were fulfilled. The best of the portfolios constructed by emergent self-organizing value maps outperformed the S&P500 Index by about 12% based on two months of out-of-sample testing.
Keywords: Clustering
|
|
|
|
Kong, A. et al, 2008. How Predictable are Compnents of the Aggregate Market Portfolio?.
Abstract: In contrast to the voluminous literature on stock return predictability that focuses on the aggregate market portfolio, we analyze return predictability for a variety of components of the aggregate market, including portfolios sorted on industries, size, and book-to-market. Considering 14 economic variables from Goyal and Welch (2008) and 33 lagged industry returns from Hong, Torous, and Valkanov (2007) as predictors, we find that returns for certain component portfolios are substantially more predictable using both in-sample and out-of-sample tests. Construction, textiles, apparel, furniture, printing, automobiles, and manufacturing are among the industry sector portfolios exhibiting the greatest degree of predictability. In addition, portfolios of small-cap and high book-to-market firms typically display greater predictability. Finally, we explore economic explanations for the differences in predictability across component portfolios. Factors such as sensitivity to macroeconomic fundamentals and risk as well as industry concentration and capitalization help to account for these differences.
Keywords: Clustering; FinanceGeneral
|
|
|
|
Pan, W. & Shen, X., 2007. Penalized Model-Based Clustering with Application to Variable Selection, Journal of Machine Learning Research, 8, p. 1145–1164.
Keywords: Clustering; DimReduction
|
|
