Nonparametric and semiparametric response surface methodology: a review of designs, models and optimization techniques

Laura Villanova, Rob J Hyndman, Kate Smith-Miles, Irene Poli Abstract: Since the introduction of Response Surface Methodology in the 1950s, there have been many developments with the aim of expanding the range of applications of the methodology. Various new design, modeling and optimization techniques have been introduced for coping with unknown input-output relationships, costly or time-consuming experimental studies and irregular experimental regions (e.g., non-cubic or non-spherical regions induced by constraints in the input variables). Such developments may involve many different research areas simultaneously (e.g., the statistical design of the experiments, multivariate modeling, and multi-objective optimization). Experts in various research fields have

(More)…

 

Coherent mortality forecasting using functional time series

Published on 11 October 2013 in Talks

A talk given today at Macquarie University, Sydney.

 

Forecasting hierarchical time series

Published on 10 October 2013 in Talks

Talk given at University of Sydney today.

 

R tools for hierarchical time series

Published on 4 July 2013 in Talks

Talk given at EURO/INFORMS, Rome, 1 July 2013 And at UseR! 2013, Albacete, Spain, 10 July 2013.

 

Forecasting without forecasters

Published on 25 June 2013 in Talks

A keynote talk given at the International Symposium on Forecasting, Seoul, South Korea. 25 June 2013.