Page 28 - Contributed Paper Session (CPS) - Volume 2

CPS1408 Caston S. et al.

Figure 2

4. Conclusion
Four models considered in this study were GAMs and AQR models with
and without interactions. The AQR model with pairwise interactions was found
to be the best fitting model. The forecasts from the four models were then
combined using an algorithm based on the pinball loss (convex combination
model) and also using quantile regression averaging (QRA). The AQR model
with interactions was then compared with the convex combination and QRA
models and the QRA model gave the most accurate forecasts. The QRA model
had the smallest prediction interval normalised average width and prediction
interval normalised average deviation.

References
1. Bien, J.; Taylor, J.; Tibshirani, R. A lasso for hierarchical interactions. The
Annals of Statistics. 2013, 41(3), 1111-1141.
https://arxiv.org/ct?url=http3A2F2Fdx.doi. org2F10252E12142F13-
AOS1096&v=3d4226e4
2. Bien, J.; Tibshirani, R. R package “hierNet”, version 1.6. 2015,
https://cran.r-project.org/ web/packages/hierNet/hierNet.pdf (Accessed
22 May 2017).
3. Fasiolo, M.; Goude, Y.; Nedellec, R.; Wood, S.N. Fast calibrated additive
quantile regression. 2017. Available
online:https://github.com/mfasiolo/qgam/blob/master/draftqgam. pdf
(Accessed on 13 March 2017).
4. Gaillard, P.; Goude, Y.; Nedellec, R. Additive models and robust
aggregation for GEFcom2014 probabilistic electric load and electricity

17 | I S I W S C 2 0 1 9