Browsing by Author "Olatayo, T. O."

Now showing 1 - 3 of 3
  • Item
    Multivariate Time Series Analysis on the Prices of Staple Foodstuffs in Kwara State, Nigeria
    (Journal of Science, Technology and Mathematics Education (JOSTMED), Federal University of Technology, Minna, 2016-12-15) Afolayan, R. B.; Yahya, W. B.; Garba, M. K.; Adenuga, A. A.; Olatayo, T. O.
    Due to supplementary, complementary and substitute relationship between staple foodstuffs, the prices of one or more staple foodstuffs tend to influence and could be used to predict the prices of some others. This study was therefore aimed at establishing the co-movement between the prices of some major stable foodstuffs - Rice, Maize, Garri, Millet, Guinea-Corn and Beans - in Kwara State, Nigeria. Multivariate time series models were fitted to data on monthly prices of Rice, Maize, Garri, Millet, Guinea-Corn and Beans over a period of twelve years (from January 2000 to December 2012). The cointegration relations among the prices were established by applying Johansen’s cointegration tests. As a result, appropriate Vector Error Correction (VEC) model was fitted to the data. The unit root test for stationarity in the series reveals that all the series were non-stationary but they were only made to be stationary at first difference. The results from the analysis showed that there exist short term adjustments and long-term dynamics among the prices of Rice, Maize, Garri, Millet, Guinea-Corn and Beans in Nigeria over the study period. Further results showed that a Vector Error Correction (VEC) model of lag two with one cointegration equation best fits the data. The forecasting accuracy of the fitted model was determined by out-of-sample forecasts of the future prices of the selected staple foodstuffs. Suitable model’s assessment criteria such as root mean square error, mean absolute error and the like were employed to determine the efficiency of the fitted model. The data employed for the study were collected from the Kwara State office of National Bureau of Statistics, Nigeria. All analyses were performed in the environment of R Statistical package.
  • Item
    On Statistical Analysis and Modeling of Rare Events with Autoregressive integrated Moving Average (Arima) Model.
    (Journal of the Nigerian Association of Mathematical Physics, 2014-11) Olatayo, T. O.; Mayor, Andrew; Alabi, O. O.; Afolayan, Razaq
    Forecasting methods to produced numerical estimates range from relative simple techniques to complex and sophisticated techniques are discussed in this paper. Among forecasting methods were extrapolative or projective techniques i.e. moving averages and exponential smoothing. A moving average is a trend method wherein each point of a moving average of a time series is the arithmetic mean of a number of consecutive observations of the series. The number of observations in the moving average computation is chosen to minimize the effect of seasonality or other disturbances in the series [l-4]. Exponential smoothing is a flexible trend whereby past data observations arc given different weights in computing the forecast. It has the advantages of providing a simple up-to-date forecast where the new forecast is equal to the previous one plus some stated proportion of the previous periods of a forecasting error. Exponential smoothing methods are adaptable to adjustment to include trend and seasonal projections with adaptive types of optimum weighting procedures. Exponential smoothing methods are used to forecast large numbers of items. Time series decomposition methods are widely used to identify the systematic components of a time scries, trend cycle and seasonal pattern and the non systematic or random component. The seasonal Pattern is identified by first determining the seasonal indexes for each month or quarter of year and in turn these patterns arc projected ahead. The cyclical forecast may be prepared by other systematic projection or by economic judgment. Non systematic or irregular variation is usually assumed zero in a forecast but irregular adjustments may be needed for an anticipated stoppage in production or some other casual factors in the time period of the forecast. A highly analytical method for measuring seasonal fluctuations is called census I l or X-l l variant. In Regression models express the past relationships among the item being forecasted. These models are useful when adequate history of data are available on the major factors associated with variations in the item being forecasted [5,6].
  • Item
    Statistical Modeling and Prediction of Rainfall Time Series Data.
    (Journal of the Nigerian Association of Mathematical Physics, 2014-07) Olatayo, T. O.; Taiwo, A. I.; Afolayan, Razaq
    Climate and rainfall are highly non-linear and complicated phenomena, which requires classical , modern and detailed models to obtain accurate prediction. In this paper, we present tools for modelling and predicting the behavioural pattern in rainfall phenomena based on past observations. The paper introduces three fundamentally different approaches for designing a model , the statistical method based on autoregressive Integrated Moving Average(ARIMA), the emerging fuzzy time series(FTS) model and non-parametric method(Theil's regression). In order to evaluate prediction efficiency, we made use of 31 years annual rainfall data from year 1982 to 2012 of Ibadan, Oyo state, Nigeria. The Fuzzy time series model has its universe of discourse divided into 13 intervals and the interval with the largest number of rainfall data is divided into 4 sub intervals of equal length. Three rules were used to determine if the forecast value under FTS is upward 0.75 -point, middle or downward 0.25-point. ARIMA(1,2,1) was used to derive the weights and the regression coefficients, while the Theil's regression was used to fit a linear model. The performance of the model was evaluated using mean squared forecast error(MAE) and root mean square forecast error (RMSE) and coefficient of determination. The study reveals that FTS model can be used as an appropriate forecasting tool to predict the rainfall, since it outperformed the ARIMA and Theil's models.