Abstract: Batch implementations of artificial neural network(ANN) are inefficient when used in an online setting because they must be retrained from scratch every time the training set is modified.In order to reduce the expense of ANN training,we have developed a dynamic neural network(DNN) modeling method for online time series prediction.Dynamic neural network modeling that efficiently updates a trained static neural network whenever a sample is added to the training set.The updated ANN function is identical to that produced by a batch algorithm.Application of DNN in an online scenario is presented.In the scenario,numerical experiments indicate that DNN is faster than batch ANN algorithms obviously.