網路拍賣結標之影響因素：以鬼滅之刃模型為例

Abstract

研究目的: (1)觀察拍賣網站上的商品資訊對結標價格的影響。 (2)了解買賣家的拍賣行為對商品結標價格的影響。 (3)觀察不同區間的結標價格受變數影響的差異性。

過程及方法： 本研究依據主要Yahoo!日本拍賣-Buyee資料，統計日本消費者的拍賣購物習慣，以分量迴歸模型（Quantile Regression），探討各項變數對於結標價格的影響，研究分為結標價格低、中、高(10-50%為低價位區間，50%以上至75%為中價位區間，75%以上至90%為高價位區間)，對這三個部分進行分量迴歸（Quantile Regression，簡稱 QR）

結果： 在不同價位下，賣家可透過對影響投標者的因素進行考慮，做出對應的行銷策略，低價位可著重於給投標者足夠時間考慮、成交後的服務，以及自身評價;中高價位可提供跳過拍賣時間直接購買價格，使拍賣者在計算利益後更容易去投標。

Research Purposes (1) Observe the impact of product information on auction websites on the bid price. (2) Understand the influence of buyers and sellers' auction behavior on the bid price of commodities. (3) Observe the difference of the bid closing prices in different intervals affected by the variables. Process and method Based on the main Yahoo! Japan Auction-Buyee data, this study counts the auction shopping habits of Japanese consumers, and uses Quantile Regression to explore the impact of various variables on the bid price. The research is divided into low bid price and medium bid price. ,High (10-50% is the low price range, 50% to 75% is the medium price range, 75% to 90% is the high price range), perform Quantile Regression (QR for short) on these three parts . Result At different price points, sellers can make corresponding marketing strategies by considering the factors affecting bidders. Low price points can focus on giving bidders enough time to consider, post-closing services, and self-evaluation; mid-to-high price points can provide Skip the auction time and purchase the price directly, making it easier for the auctioneer to bid after calculating the profit.