Determination of a better non-linear mathematical model with trigonometric sinusoidal behaviour for the pricing of local rice in Nigeria market

Abstract

The intention of the study is to determine from a list of non-linear formulated model the best that can approximate to the exact/ real life data gathered from price of local Rice in Nigeria. This is obtained/ established in the model validation section. However, the outcome of the model validation upheld that, the purely sinusoidal model performed better with a minimal average difference error of 133 unit as compared to other models. Similarly, there were several highly non-linear models that was tested with the hope of high performance, but failed. Hence, this result showed that financial variables due to their stochastic and oscillating nature can be better modelled using the sine functions and functional.