Academic Journal of Business & Management, 2020, 2(5); doi: 10.25236/AJBM.2020.020512.
Beifen Zhou1, Jinbiao Xia1, Xinru Cheng1, Binbin Xu1, Yiliang Qiao2, Xiang Wang3 and Lan Guo4
1 School of Information Science and Engineering, Guilin University of Technology, Guilin 541004, China
2 School of Mathematics and Statistics, Qilu University of Technology, Ji’nan 250353, Shandong, China
3 School of Electrical Engineering and Automation, Heilongjiang University of Science and Technology, Harbin 150027, Heilongjiang, China
4 Xi'an College of Economics and Humanities, Shanghai International Studies University, Shanghai 200083, China
Firstly, this paper obtains the SKC from July 1, 2018 to October 1, 2018 and ranks the top 50 in cumulative sales through data preprocessing. By analyzing the four holidays in 2018 and other related factors, five factors are comprehensively selected, including the actual average cost price, label price, product sales characteristics, inventory information, holiday discount, and obtains the quantitative value of each factor;The factors affecting the sales volume are calculated by the linear regression model, and then the maximum influencing factors are obtained through the linear regression analysis,The inventory information has the lowest impact on sales volume. Secondly, through data preprocessing, we get the top ten categories with the historical sales time from June 1, 2019 to October 1, 2019, and find the monthly sales volume of these sub categories from January to September, 2019, and use this month's sales volume as the original forecast data; then analyze the original data,The results show that the sales volume has a cyclical trend of rising and falling with the change of time series, so a time series forecasting model with quadratic exponential smoothing is established. Finally, the original data is brought into the model to predict the sales volume of each month in the next three months, and the predicted results of each month are obtained, and the results are brought into the MAPE formula to calculate the MAPE of the predicted value of each month in the next three months
Data preprocessing, Multiple linear regression model, Quadratic exponential smoothing time series forecasting model
Beifen Zhou, Jinbiao Xia, Xinru Cheng, Binbin Xu, Yiliang Qiao, Xiang Wang and Lan Guo. Demand Forecast of New Retail Target Products based on Multiple Linear Regression. Academic Journal of Business & Management (2020) Vol. 2, Issue 5: 114-119. https://doi.org/10.25236/AJBM.2020.020512.
[1] Yang Yanxiao, Chen Yanning, Kang Suoqian, Li Rui, Yu Xinle. Analysis of tourism income sources and influencing factors based on multiple linear regression model [J]. Rural economy and technology, 2020, 31 (07): 118-120
[2] Xiao Yunmei, Liu Qiong, Qi Shuang, Liu Yi. Prediction of cold chain demand of typical agricultural products in Changsha Zhuzhou Xiangtan region based on quadratic exponential smoothing method [J]. Rural economy and science and technology, 2019, 30 (11): 158-160
[3] Xiao Yang. Coal price prediction model based on co integration theory and artificial neural network [D]. Xi'an University of science and technology, 2016