摘要 |
PROBLEM TO BE SOLVED: To provide a demand prediction apparatus which improves prediction accuracy of purchase quantity of seasonal commodities.SOLUTION: The demand prediction apparatus includes: a storage that stores a previous purchase quantity of each commodity and variable factor information in which variable factors causing variation of the purchase quantity are digitized as an actual data; reception means that receives input of a reference date as a target of demand prediction; regression coefficient calculation means that calculates a regression coefficient in which the variable factor information represents the degree of contribution to the variation of the purchase quantity, based on first actual data for a predetermined period closest to the reference date and second actual data for a predetermined period from a date of previous year corresponding to the reference date toward the future, from the actual data stored in the storage; and predicted purchase quantity calculation means that calculates purchase quantity of each commodity purchased on the reference date as a predicted purchase quantity by using the regression coefficient calculated by the regression coefficient calculation means and the variable factor information on the reference date. |