摘要
构建加入周末系数的Bass模型，得到改进Bass模型；以消费者决策过程理论为指导，首次将网络搜索和电影特征信息两类变量为自变量，以改进Bass模型参数为因变量，建立多元回归模型；结合改进Bass模型与参数回归模型得到联合预测模型。将2013—2016年国产电影分为训练集和测试集两部分；用训练集估计得到回归模型参数，用测试集检验联合预测模型的有效性。结果显示，提出的联合预测模型能较好地预测电影日需求，并对预测结果进行分析讨论。
An improved Bass model which guidance of consumer decision-making process added a weekend coefficient was proposed. Under a theory, a multiple parameters regression model was established by taking the improved Bass model parameters as dependent variables and taking two kinds of variables including network search and films characteristics as independent variables. By combining the improved Bass model and parameters regression model, a conjoint forecasting model was obtained. Samples consisting of film showed from 2013 to 2016 in China were divided into two parts：training set and test set. The training set was used to estimate parameters of the regression model and the test set was used to test effectiveness of the conjoint forecasting model. The test shows that the conjoint forecasting model is effective result is also analyzed and discussed. to forecast films daily demand. Finally, the
作者
唐中君
王美月
禹海波
吴凡
TANG Zhong-jun;WANG Mei-yue;YU Hai-bo;WU Fan(Research Base of Beijing Modern Manufacturing Development,School of Economics and Management,Beijing University of Technology,Beijing 100124,China)
出处
《工业工程与管理》
CSSCI
北大核心
2018年第4期16-22,29,共8页
Industrial Engineering and Management
基金
国家自然科学基金资助项目(71672004)
关键词
联合预测
日需求规律
BASS模型
参数回归模型
网络搜索
conjoint forecasting
day demand patterns
Bass model
parameters regression model
network search data
