Jungsik Hong

A Parameter Estimation of Bass Diffusion Model by the Hybrid of NLS and OLS

Department of Business Administration, Chungnam National UniversityThe Bass model is a cornerstone in diffusion theory which is used for forecasting demand of durables or new services. Three well-known estimation methods for parameters of the Bass model are Ordinary Least Square (OLS), Maximum Likelihood Estimator (MLE), Nonlinear Least Square (NLS). In this paper, a hybrid method incorporating OLS and NLS is presented and it’s performance is analyzed and compared with OLS and NLS by using simulation data and empirical data. The results show that NLS has the best performance in terms of accuracy and our hybrid method has the best performance in terms of stability. Specifically, hybrid method has better performance with less data. This result means much in practical aspect because the avaliable data is little when a diffusion model is used for forecasting demand of a new product.

Easy, reliable method for mid-term demand forecasting based on the Bass model: A hybrid approach of NLS and OLS

For mid-term demand forecasting, the accuracy, stability, and ease of use of the forecasting method are considered important user requirements. We propose a new forecasting method using linearization of the hazard rate formula of the Bass model. In the proposal, reduced non-linear least square method is used to determine the market potential estimate, after the estimates for the coefficient of innovation and the coefficient of imitation are obtained by using ordinary least square method with the new linearization of the Bass model. Validations of 29 real data sets and 36 simulation data sets show that the proposed method is accurate and stable. Considering the user requirements, our method could be suitable for mid-term forecasting based on the Bass model. It has high forecasting accuracy and superior stability, is easy to understand, and can be programmed using software such as MS Excel and Matlab.

Comparison of the Bass Model and the Logistic Model from the Point of the Diffusion Theory

The logistic model and the Bass model have diverse names and formulae in diffusion theory. This diversity makes users or readers confused while it also contributes to the flexibility of modeling. The method of handling the integration constant, which is generated in process of deriving the closed form solution of the differential equation for a diffusion model, results in two different `actual` models. We rename the actual four models and propose the usage of the models with respect to the purpose of model applications. The application purpose would be the explanation of historical diffusion pattern or the forecasting of future demand. Empirical validation with 86 historical diffusion data shows that misuse of the models can draw improper conclusions for the explanation of historical diffusion pattern.

A hybrid decision tree algorithm for mixed numeric and categorical data in regression analysis

We propose a simple but efficient hybrid regression algorithm using decision tree.The proposed algorithm significantly improves combined regression algorithms.The proposed algorithm can consider the nonlinearity for categorical variables.The proposed hybrid algorithm does not increase computation cost. In many real world problems, the collected data are not always numeric; rather, the data can include categorical variables. Inclusion of different types of variables may lead to complications in regression analysis. Many regression algorithms such as linear regression, support vector regression, and neural networks that train parameters of a model to identify relations between input and output variables, can easily process numeric variables; however, there are additional considerations for categorical variables. On the other hand, a decision tree algorithm estimates a target based on the specified rules; therefore, it can support categorical variables as well as numeric variables. Using this property, a new hybrid model combining a decision tree with another regression algorithm is proposed to analyze mixed data. In the proposed model, the portions explained by categorical variables in target values are estimated by the decision tree and the remaining parts are predicted by any regression algorithm trained by numerical variables. The proposed algorithm was evaluated using 12 datasets selected from real decision problems, and it was confirmed that the proposed algorithm achieved better or comparable accuracy than the comparison methods including the M5 decision tree and the evolutionary tree. In addition, the new hybrid method does not significantly increase computational complexity, even though it builds two separate models, which is an advantage that is in contrast with the M5 decision tree and the evolutionary tree.

Box Office Forecasting considering Competitive Environment and Word-of-Mouth in Social Networks: A Case Study of Korean Film Market

Accurate box office forecasting models are developed by considering competition and word-of-mouth (WOM) effects in addition to screening-related information. Nationality, genre, ratings, and distributors of motion pictures running concurrently with the target motion picture are used to describe the competition, whereas the numbers of informative, positive, and negative mentions posted on social network services (SNS) are used to gauge the atmosphere spread by WOM. Among these candidate variables, only significant variables are selected by genetic algorithm (GA), based on which machine learning algorithms are trained to build forecasting models. The forecasts are combined to improve forecasting performance. Experimental results on the Korean film market show that the forecasting accuracy in early screening periods can be significantly improved by considering competition. In addition, WOM has a stronger influence on total box office forecasting. Considering both competition and WOM improves forecasting performance to a larger extent than when only one of them is considered.