Original Research Article
Year: 2015 | Month: June | Volume: 2 | Issue: 6 | Pages: 372-378
Forecasting Share Prices of Banks of Colombo Stock Exchange
W.G. S. Konarasinghe1, N. R. Abeynayake2, L.H.P.Gunaratne3
1Postgraduate Institute of Agriculture, University of Peradeniya, Sri Lanka,
2Faculty of Agriculture and Plantation Management, Wayamba University of Sri Lanka, Makandura, Gonawila (NWP), Sri Lanka,
3Department of Agricultural Economics and Business Management, Faculty of Agriculture, University of Peradeniya, Peradeniya, Sri Lanka
Corresponding Author: samanthi
ABSTRACT
Scientific forecasting plays a vital role in financial markets for many decades. Forecasting methods of financial market were broadly classified into fundamental analysis and technical analysis. Fundamental analysis involves analyzing economic and firm specific factors while technical analysis is concerned with the price movements and trading volume in the market. Both approaches have been applied in forecasting share prices and stock returns. However, only a limited number of studies have been done on Sri Lankan share market. Existing forecasts made mainly based on Capital Asset Pricing Model has been proved unreliable in the Sri Lankan context. This study aimed to test the suitability of linear models, non-linear models and univariate causal models for forecasting share prices based on the data during 1994-2011 on randomly selected three banks listed in Colombo Stock Exchange (CSE). Findings suggest that univariate causal models were the most suitable in forecasting share prices of individual companies listed in CSE.
Keywords: linear model, non- linear model