Search

Your search keyword '"Abeynayake, N. R."' showing total 18 results
Start Over Author "Abeynayake, N. R."
18 results on '"Abeynayake, N. R."'

11. Unobserved Components Model for Forecasting Sugarcane Production in Sri Lanka.

Author

Ariyawansha, B. D. S. K., Abeynayake, N. R., and Sivananthawerl, T.

Subjects
SUGARCANE, BOX-Jenkins forecasting, STATISTICAL smoothing, AKAIKE information criterion, INFORMATION theory
Abstract

Unobserved Components Model (UCM) is a structural time series model and it can decompose the response series into latent components, such as trend, cycle and seasonal effects and linear and nonlinear regression effects. The UCM combines the capabilities of Autoregressive Integrated Moving Average (ARIMA) model with interpretability of smoothing models. This study was carried out to forecast sugarcane production in Sri Lanka using UCM model. The best fitting model was selected based on Akaike information criterion (AIC) and Bayesian Information Criterion (BIC), followed by residual analysis. The selected model was used to make sample period forecasts (From 1979 - 2013) and post sample period forecasts (From 2014 to 2018). Forecasting accuracy of model was evaluated using the Mean Absolute Percentage Error (MAPE). Linear trend model (adj. R²=77 %) with zero variance slope and two cycles was selected as best among the tested UCM models for cane production data. MAPE was 10.56 % for sample period forecasts and 4.01 % for post-sample period forecasts. Predicted cane production for year 2019 was 813,888 ± 293,891 tons. [ABSTRACT FROM AUTHOR]

Published
2020
Full Text
View/download PDF

13. TIME SERIES PATTERNS OF SRI LANKAN STOCK RETURNS

Author

Konarasinghe W G S. and Abeynayake N R.

Subjects
Stationary, ANOVA, ARIMA, Spectral Analysis
Abstract

Predicting the future was one of the strongest desires of man. The actualization of the desires historically began with astrology paying the path for scientific predictions and forecasting. Scientific forecasting in financial markets also has a long history going back to 1950’s. Literature shows several attempts in modeling share returns in Sri Lankan context, but outcomes of the studies were not satisfactory. Colombo Stock Exchange (CSE) is a population containing 20 mutually exclusive subsets (sectors). Elements of these subsets are homogeneous in nature of business. As such, there may be similarities in market performances within groups and similar forecasting techniques may successful in forecasting. The present study was focused on sector-wise pattern recognition of returns and identifying suitable time series forecasting techniques. A random sample of four business sectors was selected and required data were obtained from the CSE data library, 2011. Box- Plots of returns and one–way ANOVA was used for comparison of returns within and between sectors. Results did not give evidence for significant differences in mean returns. Time series plots of returns of individual companies as well as sectors showed no trends. Stationary of the series were tested by Auto Correlation Functions (ACF’s) and found all the series stationary. Based on the results it was concluded that company returns, as well as sector returns of CSE, were stationary. It is recommended to test ARIMA models and Spectral Analysis in forecasting returns of the Sri Lankan share market., {"references":["Konarasinghe, W .G. S., & Abeynayake, N. R. (2014). Time series patterns of Sri Lankan stock returns. Proceedings of the 11th International Conference on Business Management, Phd Consortium, Faculty of Management studies and Commerce, University of Sri Jayewardenepura. 78-95."]}

Published
2014
Full Text
View/download PDF

18. Robustness of neighbour balanced complete block designs against missing observation(s).

Author

Abeynayake, N. R., Jaggi, Seema, and Varghese, Cini

Subjects
ROBUST control, MISSING data (Statistics), BLOCK designs, EIGENVALUES, ESTIMATION theory
Abstract

A block design with neighbour effect(s) is said to be neighbour balanced if every treatment has every other treatment appearing constant number of times as neighbour(s). These designs are used when the treatment applied to one experimental plot may affect the response on neighbouring plot(s) besides the response to which it is applied. Neighbour-balanced designs ensure that no treatment is unduly disadvantaged by its neighbour(s). However, there is a possibility that some of the observations could become unavailable for analysis. In this paper, we examine the robustness of neighbour balanced complete block designs when specific observations are missing. The information matrix for direct treatment effects of the resultant design (one-sided neighbour effects) after missing of an observation from a block is derived and the efficiency of resulting design is investigated. Robustness of neighbour balanced complete block design has also been investigated against missing of more than one observation. The efficiencies are found to be quite high indicating the designs to be fairly robust against missing observations. [ABSTRACT FROM AUTHOR]

Published
2011
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results