政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/129727

政大典藏 > College of Informatics > Department of Computer Science > Books & Chapters in Books > Item 140.119/129727

Please use this identifier to cite or link to this item: https://nccur.lib.nccu.edu.tw/handle/140.119/129727

Title:	A Study of Models for Forecasting E-Commerce Sales During a Price War in the Medical Product Industry
Authors:	謝佩璇 Hsieh, P. H.
Contributors:	資科系
Keywords:	E-commerce ;Price war ;Sales forecasting; Inventory plan
Date:	2019-06
Issue Date:	2020-05-25 10:15:00 (UTC+8)
Abstract:	When faced with a price war, the accuracy of forecasting sales in e-commerce greatly influences an enterprise’s or a retailer’s merchandise inventory strategies. When faced with a price war, an enterprise might obtain certain consumption patterns by analyzing previous sales data. This case study research was conducted in collaboration with a medical product company to explore which of the various forecasting models can better inform a company’s inventory plan. The study used the company’s data from Amazon.com regarding sales volume, number of views, company ranking, etc. between February 7 2016 and March 28 of 2018. Three potential methods of data mining were selected from the literature: the exponential smoothing method, the linear trend method, and the seasonal variation method. Of these, the most suitable was identified for price war situations to forecast the sales volume for April 2018 and to provide concrete information for the company’s inventory plan. The results showed that the seasonal variation method is more suitable than the other two sales forecasting methods. To obtain a more accurate sales forecast during a price war, the seasonal variation method is recommended to be used in the following approaches: Adjust the seasonal index by using a simple moving average. Remove the seasonal index from the sales volume, and conduct a regression analysis using the data within the last month. The resulting predicted value (with the seasonal index removed) should be multiplied by each period’s corresponding weighted moving average to obtain a more accurate sales forecast during a price war.
Relation:	HCI in Business, Government and Organizations. eCommerce and Consumer Behavior, Springer, pp.3-21
Data Type:	專書篇章
DOI:	10.1007/978-3-030-22335-9_1 https://doi.org/10.1007/978-3-030-22335-9_1
Appears in Collections:	[Department of Computer Science ] Books & Chapters in Books

Files in This Item:

File	Description	Size	Format
198.pdf		591Kb	Adobe PDF2	278	View/Open

社群 sharing

Loading...