Determination of starting values in estimating growth curves by using non-linear least squares

비선형 최소자승법을 이용한 성장곡선 모형의 매개변수 추정시 초기값 설정 방법에 관한 연구

Growth curves including Logistic and Gompertz functions are widely used in forecasting the market demand. To estimated the parameters of those functions, we use the non-linear least square method. However, it is difficult to set up the starting points for each parameter. If a wrong starting point is selected, the result reveals the local optimum or does not converge to a certain value. The purpose of this paper is to resolve the problem of selecting a starting point. Especially, rescaling the market data using the national economic index make it possible to figure out the range of parameters and to utilize the grid search method. Applications to some real data are also included.