Browsing by Author "Cooray, T.M.J.A."

Now showing 1 - 3 of 3

Autoregressive Modelling Approach to Forecasting Paddy Yield
(University of Kelaniya, 2005) Cooray, T.M.J.A.
This study was carried out to evaluate the alternative univariate time series modelling approaches for main agricultural output in Sri Lanka. For the purpose of the study total production of paddy were used. Those values were obtained from the Annual Bulletin, published by Central Bank of Sri Lanka. Box-Jenkins ARIMA methodology is used to assess Autoregressive model presented in this research. In order to compare how well implementations of the two techniques work, these methods were applied to forecast three values (last three) of each fitted model. Forecast errors were used to assess mean error (ME) and mean of the absolute percentage error (MAPE) from each data series. Those MAPE values are 11% and 6% corresponding to the ARIMA modelling and Auto Regression Approach respectively. It can be considered that autoregression method can play an important role of time series modelling; especially it is a useful method to assess an appropriate model if the original series do not show seasonality or no strong evidence for stationary.
Model for Tourism Forecasting and Policy in Complex Situations
(University of Kelaniya, 2005) Cooray, T.M.J.A.
Information about the current and forecast levels of tourism and its contribution to the economy is important for policymaking by businesses and governments. Traditional forecasting methods can provide reasonable forecasts in the context of predicable changes. However, forecasting becomes problematic in the context of both predictable changes and less predictable domestic or international shocks. This paper demonstrates the ways in which an integrated model, combining traditional forecasting methods and quantifiable scenario forecasts, can be used to examine complex combinations of events. The model is applied to Sri Lanka’s tourism indicators, which provide a picture of tourism in the Sri Lankan economy, with traditional forecasting methods and quantifiable scenario forecasts. Results are provided for a combination of changes in relative exchange rates, income of major origin countries and a positive shock to tourism demand, to demonstrate the integrated model's ability to take account of the complex multiple events that affect tourism destinations.
VAR (Vector Auto Regression) approach for analysis of econometric series
(Faculty of Social Sciences, University of Kelaniya, 2009) Cooray, T.M.J.A.