AIC microseismic first break pickup
no vote
AIC (Akaike Information Criterion) is a statistical model selection method used to compare the fitting capabilities of different models. Its principle is based on information theory and provides a method for balancing model fit and model complexity. The AIC method was proposed by the Japanese statistician Hirotsugu Akaike in 1974. The basic idea is that the better the fitting ability of the model under given data, the smaller the AIC value of the model. The calculation formula for the AIC method is as follows: AIC=- 2 ln (L)+2 k, where L is the maximum likelihood function value of the model and k is the number of parameters of the model. The smaller the AIC value, the better the model. Because with the same degree of fit, models with smaller AIC values have fewer parameters, which means they are more simple. Therefore, the AIC method can prevent the model from over fitting data and improve the generalization ability of the model. The AIC method is not only suitable for linear models, but also for generalized linear models, nonlinear models, and time series models. It has been widely used in many fields, such as economics, ecology, physics, psychology, and biostatistics.