Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/21333
Title: | 以巴斯擴散模型預測新產品銷售模式:以台灣超市為例 New Product Demand Forecasting Based on Bass Diffusion Model – In case of a Chain Supermarket in Taiwan |
Authors: | Te-An Wang 王得安 |
Advisor: | 任立中(Li-Chung Jen) |
Keyword: | 超市,新產品預測,Bass 擴散模型,非線性回歸,彷彿無相關回歸模型, Supermarket,New product forecasting,Bass Diffusion Model,Nonlinear Least Squares,Seemingly Unrelated Regression Model, |
Publication Year : | 2019 |
Degree: | 碩士 |
Abstract: | 現代消費者購買習慣變化快速,面對動態市場,零售通路最大的難題之一即為銷量預測,因為銷售的數量將直接影響到庫存管理、業務資源配置、行銷費用安排,而當企業能降低庫存、協調人力資源、準確評估行銷費用時,甚至能幫助公司的股價市值扶搖直上。更確切來說,當我們建立了良好的銷售預測模型,可以從未來產品的規劃中推估公司的營運銷售額成長趨勢、降低企業營運風險、提升資源運用效率。
作為眾多市場分析模型中的一種,Bass 擴散模型最初的設計主要應用於新上市的消費性耐用品銷售量預測,近年來也常見論文應用於非耐久品消費品的預測,比如電影。本次研究則嘗試以 Bass 擴散模型搭配彷彿無相關迴歸模型預測跨種類產品之銷售量,雖然不同產品於銷售價格、銷售數量、淡旺季、促銷敏感度差異大,但此研究針對影響銷售數量的重要變數進行轉換,找出解釋力較佳的模型,並且檢驗其預測效度。 本研究資料之時空背景為 2009~2010 年,根據經濟部統計資料,公有傳統市場營業額於 1999 年約為 965 億元,民國 2010 年降至 800 億元,而到了民國 2013 年跌至 698 億元,代表著這段期間現代超市逐漸取代傳統市場、消費者購買習慣改變。此外,在超市競爭日益激烈的情況下,通路的產品豐富度、地點的方便性、優惠的價格一直都是吸引客人前往購買的關鍵點,而良好的銷量預測模型可以優化庫存空間的使用、降低管銷成本,幫助上述關鍵因素。本研究將以台灣超市通路 – 松青超市的實際銷售資料作為分析來源,透過創新產品銷售量預測模型預測不同品類間的產品銷量,提供各大零售通路之預測參考。 The tastes and preferences of the current customer are changing rapidly, one of the biggest problems for retail channels is to do a sales forecast in this dynamic market. Sales forecasting can play a major role in driving company's stock prices by optimizing inventory management, human resource allocation, marketing budget allocation, etc. More specifically, with a precise new product demand forecasting model, we can predict a company’s revenue growth from product plans to reduce operational risk and improve operational efficiencies. As one of the most popular forecasting models, the Bass diffusion model was originally designed to predict the demand for consumer durable goods. In recent years, it is also applied to non-durable goods such as movie in many theses. This study aims to predict the sales volume for non-durable and cross-category products by using Bass diffusion model and Seemingly unrelated regression model (SUR). Although the price range, sales volume, product life cycle, and price sensitivity might be different for the products, this study will standardize the important variables affecting sales and find the best prediction model with good predictive validity. According to the Ministry of Economic Affairs, R.O.C., the public traditional market was 96.5 billion NTD in 1999, 80 billion NTD in 2010, and fell to 69.8 billion NTD in 2013. This trend shows that modern supermarket gradually replaced the traditional market and consumers' buying behavior changed. In addition, the keys to attracting potential customers are abundant products, convenience, and price. A good sales forecasting model can optimize the use of inventory space and reduce the cost of management to facilitate those key factors. Therefore, the present study used the actual sales data of Taiwan Matsusei Supermarket from 2009 to 2010 as the object of empirical analysis, predicting the sales volume of different types of products through the innovative product sales forecasting model, and expect to provide a reference for retail channels. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/21333 |
DOI: | 10.6342/NTU201902191 |
Fulltext Rights: | 未授權 |
Appears in Collections: | 國際企業學系 |
Files in This Item:
File | Size | Format | |
---|---|---|---|
ntu-108-1.pdf Restricted Access | 16.61 MB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.