Hyndsight

Thoughts on research, forecasting, statistics, and other distractions.

Topics covered

All Hyndsight posts by date

forecast v8.3 now on CRAN

14 April 2018

The latest version of the forecast package for R is now on CRAN. This is the version used in the 2nd edition of my forecasting textbook with George Athanasopoulos. So readers should now be able to replicate all examples in the book using only CRAN packages. A few new features of the forecast package may be of interest. A more complete Changelog is also available. mstl() handles multiple seasonality STL decomposition was designed to handle a single type of seasonality, but modern data often involves several seasonal periods (e.

A brief history of time series forecasting competitions

11 April 2018

forecasting, reproducible research, time series, R

Prediction competitions are now so widespread that it is often forgotten how controversial they were when first held, and how influential they have been over the years. To keep this exercise manageable, I will restrict attention to time series forecasting competitions — where only the history of the data is available when producing forecasts. Nottingham studies The earliest non-trivial study of time series forecast accuracy was probably by David Reid as part of his PhD at the University of Nottingham (1969).

R package for M4 Forecasting Competition

26 March 2018

forecasting, R, reproducible research, time series

The M4 forecasting competition is well under-way, and a few of my PhD students have been working on submissions. Pablo Montero-Manso, Carla Netto, and Thiyanga Talagala have made an R package containing all of the data (100,000 time series), which should make it substantially easier for other contestants to load the data into R in order to compute forecasts. Grab the package from this github repository. For more details about the M4 competition see this post or go to the M4 website.

IJF Tao Hong Award 2018

13 March 2018

prizes, forecasting, energy, IJF

Every two years, the International Journal of Forecasting awards a prize to the best paper on energy forecasting. The prize is generously funded by Professor Tao Hong. This year, we will award the prize to a paper published in the IJF during the period 2015-2016. The prize will be US$1000 plus an engraved plaque. The award committee is Rob J Hyndman, Pierre Pinson and James Mitchell. Nominations are invited from any reader of the IJF.

M4 Forecasting Competition update

22 December 2017

forecasting, R, reproducible research, time series

The official guidelines for the M4 competition have now been published, and there have been several developments since my last post on this. There is now a prize for prediction interval accuracy using a scaled version of the Mean Interval Score. If the $100(1-\alpha)$% prediction interval for time $t$ is given by $[L_{t},U_{t}]$, for $t=1,\dots,h$, then the MIS is defined as $$\frac{1}{h}\sum_{t=1}^{h} \left[ (U_t-L_t) + \frac{2}{\alpha}(L_t-Y_t)1(Y_t < L_t) + \frac{2}{\alpha}(Y_t-U_t)1(Y_t > U_t) \right] $$ where $Y_t$ is the observation at time $t$.

Data Science for Managers: May 2018

5 December 2017

conferences, data science, Monash University

For the last few years, I have been involved with running a 3-day short course on “Data Science for Managers”. We have run it twice each year since 2015, and it continues to prove very popular. We have some awesome presenters including Monash University professors Di Cook, Geoff Webb, and Kim Marriott, as well as several very experienced data scientists working in industry. The next course will be held on 8-10 May 2018.

Some new time series packages

29 November 2017

R, reproducible-research, time-series, data-science

This week I have finished preliminary versions of two new R packages for time series analysis. The first (tscompdata) contains several large collections of time series that have been used in forecasting competitions; the second (tsfeatures) is designed to compute features from univariate time series data. For now, both are only on github. I will probably submit them to CRAN after they’ve been tested by a few more people.

M4 Forecasting Competition: response from Spyros Makridakis

20 November 2017

forecasting, R, reproducible research, time series

Following my post on the M4 competition yesterday, Spyros Makridakis sent me these comments for posting here. I would like to thank Rob, my friend and co-author, for his insightful remarks concerning the upcoming M4 competition. As Rob says, the two of us have talked a great deal about competitions and I certainly agree with him about the “ideal” forecasting competition. In this reply, I will explain why I have deviated from the “ideal”, mostly for practical reasons and to ensure higher participation.

M4 Forecasting Competition

19 November 2017

forecasting, R, reproducible research, time series

The “M” competitions organized by Spyros Makridakis have had an enormous influence on the field of forecasting. They focused attention on what models produced good forecasts, rather than on the mathematical properties of those models. For that, Spyros deserves congratulations for changing the landscape of forecasting research through this series of competitions. Makridakis & Hibon, (JRSSA 1979) was the first serious attempt at a large empirical evaluation of forecast methods.

Come and work with me

14 November 2017

Monash University, jobs, acems, R, data science

I have funding for a new post-doctoral research fellow, on a 2-year contract, to work with me and Professor Kate Smith-Miles on analysing large collections of time series data. We are particularly seeking someone with a PhD in computational statistics or statistical machine learning. Desirable characteristics: Experience with time series data. Experience with R package development. Familiarity with reproducible research practices (e.g., git, rmarkdown, etc). A background in machine learning or computational statistics.