딥러닝·AI 기반 농식품 물가 변동성 분석 및 예측

Abstract

본 연구는 신선 농산물에서 가공식품·외식까지 식품 체계 전체를 대상으로, 품목군별로 물가를 어디까지 예측할 수 있는지, 그리고 그 차이를 결정하는 요인이 무엇인지를 AI·딥러닝으로 규명하고자 함. 궁극적으로는 더 정확하면서도 근거를 설명할 수 있는 물가 예측을 가능하게 하는 것을 목적으로 함.

Background of the Study ○Agri-food prices consist of the prices of many item groups, from fresh produce to processed foods and food away from home, and each group's prices are formed through a different channel. Changes in raw material costs, energy, exchange rates, and distribution margins pass through to processed food and food-away-from-home prices with a lag, and they appear first in upstream price statistics such as producer prices.

○Fresh fruit and vegetable prices, by contrast, are driven largely by supply shocks—crop conditions, regional weather, storage and shipment, and wholesale market arrivals—that monthly public data do not reveal in advance, so a single forecasting approach cannot be expected to work for every item group.

(For more information, please refer to the report.)