Study of logistic growth curve model for mobile user growth
Jin Tao, Deyong Gao
Abstract
Jin Tao, Deyong Gao
Abstract
Many Telecom Service companies need to forecast mobile user growth demand because their lead-time to supply is longer than their customers will typically wait for products. A logistic growth curve is an sigmoid curve that can be used to forecast this growth trends, so these companies adopt forecasted data for production planning. In order to construct logistic growth curve model, three phase sum method can be employed. Three phase sum method means that the whole time sequence is divided into three equal time phase, the parameters are computed according to the sum of the observed values of three time phase. But the prediction accuracy of this method is limited, the logistic curve model which can be transformed, employ ordinary least-squares principle to simply formula. The 0.618 optimal seeking method is applied to optimize Model, which adjusts key parameters of logistic growth curve for the purpose of minimizing the sum of squared residuals and better fitting actual data. The 0.618 optimal seeking method can effectively reduce the search time and increase the efficiency of fitting the data. Although the logistic model establishment for Postal and Telecommunication Services is demonstrated in this paper, this model can be applied in many fields. Example analysis for specifying these models based on the use of the logistic curve model are also provided.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Many Telecom Service companies need to forecast mobile user growth demand because their lead-time to supply is longer than their customers will typically wait for products. A logistic growth curve is an sigmoid curve that can be used to forecast this growth trends, so these companies adopt forecasted data for production planning. In order to construct logistic growth curve model, three phase sum method can be employed. Three phase sum method means that the whole time sequence is divided into three equal time phase, the parameters are computed according to the sum of the observed values of three time phase. But the prediction accuracy of this method is limited, the logistic curve model which can be transformed, employ ordinary least-squares principle to simply formula. The 0.618 optimal seeking method is applied to optimize Model, which adjusts key parameters of logistic growth curve for the purpose of minimizing the sum of squared residuals and better fitting actual data. The 0.618 optimal seeking method can effectively reduce the search time and increase the efficiency of fitting the data. Although the logistic model establishment for Postal and Telecommunication Services is demonstrated in this paper, this model can be applied in many fields. Example analysis for specifying these models based on the use of the logistic curve model are also provided.
Key concepts: Logistic function, Sigmoid function, Logistic regression, Growth curve (statistics), Curve fitting, Computer science, Ordinary least squares, Key (lock)