Study On Forecasting Second-hand Vessel Price Index Based On BP Neural Network

As to every shipping company, ships are base of survival and development. Investment on ships is a necessary economic activity for shipping companies. As a capital investment, ship owner must choose proper opportunity to purchase ships, for which this investment influence the vessel's net revenue, then influence the fleet' cash flow and revenue. Timing of second-hand vessel on investment is more feasible in forecast and operation than new-building vessel.This thesis focuses on 5-year-old's second-hand dry-bulker, because dry-bulk shipping market has the largest risk and most violent fluctuation in the international shipping market. As for relative market, second-hand dry-bulker's price change fiercely going with fluctuation of dry-bulk shipping market. Accurate prediction of second-hand dry-bulker price index (SHDBPI) can guide effectively many ship-owners to venture investment. SHDBPI fluctuates violently and gets out of trend for being influenced by many factors, which bring huge risk and choice for operator in market. Just going on that point, the thesis combines qualitative and quantitative analysis, mainly adopts BP neural network to search for laws of index and forecasting index effectively.Demand and supply of international dry-bulk shipping market,especially ,demand and supply of second-hand vessel,influence factors and evolvement and inherent laws on SHDBPI have been deeply analyzed in the thesis, which lay the groundwork to analyze dry-bulk shipping market and SHDBPI's trend. Then SHDBPI's development, algorithmic method and historical march have been analyzed, and its character also has been completely introduced and analyzed in the thesis.This thesis adopts BP neural network prediction method.BP neural network has huge capacity to non-linearity simulate, self-organize and self-study, can grasp the inherent laws of things well. The forecasted results can realize well to grasp the development trend of SHDBPI, the errors between forecasted and real results are held under 4%. Forecasting value is accurate.