The methods are described in Dokumentov, A., and Hyndman, R.J. (2014) Bivariate data with ridges: two-dimensional smoothing of mortality rates.

Authors: Alex Dokumentov and Rob J Hyndman

R Code

install.packages("smoothAPC") |

R Code

library(smoothAPC) library(demography) m <- log(fr.mort$rate$female[1:30, 150:160]) sm <- smoothAPC(m, lambdaaa = 0.2, lambdayy = 0.1, lambdaay = 0.4, effects = FALSE) plot3d(sm) |

This package is free and open source software, licensed under GPL (>= 2).

]]>

Authors: Alex Dokumentov and Rob J Hyndman

R Code

install.packages("stR") |

R Code

library(stR) co2.fit <- AutoSTR(co2) plot(co2.fit) |

This package is free and open source software, licensed under GPL (>= 2).

]]>*Journal of Computational and Graphical Statistics* (2016) to appear.

**Abstract**

Age-specific mortality rates are often disaggregated by different attributes, such as sex, state and ethnicity. Forecasting age-specific mortality rates at the national and sub-national levels plays an important role in developing social policy. However, independent forecasts of age-specific mortality rates at the sub-national levels may not add up to the forecasts at the national level. To address this issue, we consider the problem of reconciling age-specific mortality rate forecasts from the viewpoint of grouped univariate time series forecasting methods (Hyndman et al, 2011), and extend these methods to functional time series forecasting, where age is considered as a continuum. The grouped functional time series methods are used to produce point forecasts of mortality rates that are aggregated appropriately across different disaggregation factors. For evaluating forecast uncertainty, we propose a bootstrap method for reconciling interval forecasts. Using the regional age-specific mortality rates in Japan, obtained from the Japanese Mortality Database, we investigate the one- to ten-step-ahead point and interval forecast accuracies between the independent and grouped functional time series forecasting methods. The proposed methods are shown to be useful for reconciling forecasts of age-specific mortality rates at the national and sub-national levels, and they also enjoy improved forecast accuracy averaged over different disaggregation factors.

Authors: Rob J Hyndman and Nikolaos Kourentzes

R Code

# install.packages("devtools") devtools::install_github("robjhyndman/thief") |

R Code

library(thief) thief(USAccDeaths) |

This package is free and open source software, licensed under GPL (>= 2).

]]>Monday 20 June 2016

Santander, Spain

It is common practice to evaluate the strength of forecasting methods using collections of well-studied time series datasets, such as the M3 data. But how diverse are these time series, how challenging, and do they enable us to study the unique strengths and weaknesses of different forecasting methods? In this paper we propose a visualisation method for a collection of time series that enables a time series to be represented as a point in a 2-dimensional feature space. The effectiveness of different forecasting methods can be visualised easily across this space, and the diversity of the time series in an existing collection can be assessed. Noting that the M3 dataset is not as diverse as we would ideally like, this paper also proposes a method for generating new time series with controllable characteristics to fill in and spread out the instance space, making generalisations of forecasting method performance as robust as possible.

]]>*Journal of Allergy and Clinical Immunology* (2016)

**Background:** Childhood asthma is a significant public health problem and severe exacerbation can result in diminished quality of life and hospitalisation.

**Objective:** To examine the contribution of outdoor fungi to childhood and adolescent asthma hospitalisations

**Methods:** The Melbourne Air Pollen Children and Adolescent (MAPCAH) study is a case-crossover study of 644 children and adolescents (aged 2-17 years) hospitalised for asthma between September 2009 and December 2011. MAPCAH collected individual data on human rhinovirus (HRV) infection and fungal sensitisation; and daily counts of ambient concentrations of fungal spores, pollen and air pollutants. Conditional logistic regression models were used to assess associations between interquartile increases in spore counts and controlling for potential confounding and interactions.

**Results:** Exposure to *Alternaria* (aOR=1.07, 95%CI 1.03-1.11, *Leptosphaeria* (aOR=1.05, 95%CI 1.02-1.07), *Coprinus* (aOR=1.04, 95%CI 1.01-1.07), *Drechslera* (aOR=1.03, 95%CI1.00-1.05) and total spores (aOR=1.05, 95%CI 1.01-1.09) were significantly associated with child asthma hospitalisations independent of HRV infection. There were significant lagged effects up to 3-days with *Alternaria*, *Leptosphaeria*, *Cladosporium*, *Sporormiella*, *Coprinus*, and *Drechslera*. Some of these associations were significantly greater in participants with *Cladosporium* sensitisation.

**Conclusion:** Exposures to several outdoor fungal spore taxa, including some not reported in previous research, are associated with the risk of child and adolescent asthma hospitalisation, particularly in individuals who are sensitised to *Cladosporium*. We need further studies to examine cross-reactivity causing asthma exacerbations. Identifying sensitisation to multiple fungal allergens in asthmatic children could support the design and implementation of more effective strategies to prevent asthma exacerbations.

]]>

Hyndman and Athanasopoulos (2014): OTexts.org/fpp

Make sure you have a recent version of R and RStudio, and have installed the fpp and ggplot2 packages and all their dependencies.

- Introduction: Slides
- Time series visualization Slides [R code] [GDP data]
- Benchmark forecasting and time series decomposition Slides [R code]
- Exponential smoothing methods Slides [R code]
- ARIMA models Slides [R code]

- School of Statistics, Renmin University of China.
- Department of Econometrics and Business Statistics, Monash University, Australia.
- School of Mathematical Sciences, Monash University, Australia.

*International Journal of Forecasting*, 2017. to appear.

**Abstract**

It is common practice to evaluate the strength of forecasting methods using collections of well-studied time series datasets, such as the M3 data. But how diverse are these time series, how challenging, and do they enable us to study the unique strengths and weaknesses of different forecasting methods? In this paper we propose a visualisation method for a collection of time series that enables a time series to be represented as a point in a 2-dimensional instance space. The effectiveness of different forecasting methods can be visualised easily across this space, and the diversity of the time series in an existing collection can be assessed. Noting that the M3 dataset is not as diverse as we would ideally like, this paper also proposes a method for generating new time series with controllable characteristics to fill in and spread out the instance space, making generalisations of forecasting method performance as robust as possible.

]]>