Variable fractional order Nesterov accelerated gradient algorithm for rational models with missing inputs using interpolated model

An interpolated model based variable fractional order Nesterov accelerated gradient (IM-VF-NAG) algorithm is proposed to estimate the parameters of rational models with missing input data. This algorithm employs the interpolated model to impute the missing input data, making the parameter estimation more accurate. In addition, the fractional order derivatives and variable order method are used in the Nesterov accelerated gradient (NAG) algorithm, with the aim at enhancing the stability and increasing the convergence rates. To verify the efficacy of the proposed algorithm, a comprehensive simulation example is provided.