Trend Analysis and Short-Term Forecasting of Goat and Sheep Populations and their Meat Production in Sri Lanka using Single and Double Exponential Smoothing Models

Trend analysis and forecasting have become important tools in different fields. The small ruminants sub-sector is an important part of animal production in Asia as well as in Sri Lanka. The present study was conducted to identify the best trend and the time series model to forecast the goat and sheep population and their meat production in Sri Lanka. The findings of the study were based on the goat population and sheep population data during the years from 1960 to 2018 and mutton production and numbers of slaughtered goats and sheep during the years 1991 to 2017. Data were analyzed and the best model was selected according to the mean absolute percentage error (MAPE). Single Exponential Smoothing and Double Exponential Smoothing models were used to perform the short term forecasting. Among the Exponential models, the Single Exponential model was fitted only for the sheep population while the Double Exponential model was fitted for the goat population, mutton production and numbers of slaughtered goats and sheep. Sheep population, mutton production and numbers of slaughtered goat and sheep animals showed a decreasing trend while goat population showed a decreasing trend after the year 1978.