One paper about a Model-free prediction in the regression context with DNN (simple and interesting).

We propose a scalable subsampling method on estimating standard fully connected DNN. It turns out that the mean square error bound of estimation based on DNN to a target regression function can be improved under mild conditions. Moreover, it can run faster than training a single DNN on the whole dataset. In addition, we propose various methods to estimate the bias order of DNN, build confidence and prediction intervals based on DNN.

We construct bootstrap prediction intervals in the multi-step ahead prediction problem with parametric model estimator; in particular, we develop an asymptotically valid quantile prediction interval as well as a pertinent prediction interval for future values. To correct the undercoverage of prediction intervals with finite samples, we further employ predictive—as opposed to fitted—residuals in the bootstrap process.

We consider a forward bootstrap approach with non-parametric model estimator to resolve the difficulty on predicting multi-step ahead time series data. We construct a quantile prediction interval that is asymptotically valid. Moreover, after taking a debiasing technique, we can build pertinent prediction intervals in which the estimation variability is captured. (This is my first paper with my advisor Professor Politis. Thanks for his guidance!)

This paper proposes a new NoVaS prediction method based on the GARCH model. Compared to the existing NoVaS method which is motivated by ARCH model. We show the our new method is more robust and accurate.

This paper applies a so-called Molde-free prediction method—NoVaS—to do volatility forecasting of financial series. We develop a more parsimonious variant of the existing method. With systematic justification and extensive data analysis, we verify its great performance on long-term time aggregated predictions. (This is my first paper with Professor Karmakar. Thanks for his guidance!)