Forecasts are always wrong
Recently I was interviewed for the Monash Business School podcast “Thought Capital” on the topic of forecasting. You can listen to the episode here (or read the transcript).
Non-Gaussian forecasting using fable
library(tidyverse) library(tsibble) library(lubridate) library(feasts) library(fable) In my previous post about the new fable package, we saw how fable can produce forecast distributions, not just point forecasts. All my examples used Gaussian (normal) distributions, so in this post I want to show how non-Gaussian forecasting can be done. As an example, we will use eating-out expenditure in my home state of Victoria. vic_cafe <- tsibbledata::aus_retail %>% filter( State == "Victoria", Industry == "Cafes, restaurants and catering services" ) %>% select(Month, Turnover) vic_cafe %>% autoplot(Turnover) + ggtitle("Monthly turnover of Victorian cafes") Forecasting with transformations Clearly the variance is increasing with the level of the series, so we will consider modelling a Box-Cox transformation of the data.Tidy forecasting in R
The fable package for doing tidy forecasting in R is now on CRAN. Like tsibble and feasts, it is also part of the tidyverts family of packages for analysing, modelling and forecasting many related time series (stored as tsibbles). For a brief introduction to tsibbles, see this post from last month. Here we will forecast Australian tourism data by state/region and purpose. This data is stored in the tourism tsibble where Trips contains domestic visitor nights in thousands.Recent publications
- Earo Wang, Dianne Cook, Rob J Hyndman (2020) Calendar-based graphics for visualizing people's daily schedules. J Computational & Graphical Statistics, to appear. Abstract DOI pdf
- George Athanasopoulos, Puwasala Gamakumara, Anastasios Panagiotelis, Rob J Hyndman and Mohamed Affan (2020) Hierarchical forecasting. Macroeconomic forecasting in the era of big data, ed. P. Fuleky, Springer, Chapter 21, pp.689-719. Abstract DOI pdf
- Jeremy Forbes, Dianne Cook, Rob J Hyndman (2020) Spatial modelling of the two-party preferred vote in Australian federal elections: 2001-2016. Australian and New Zealand Journal of Statistics, to appear. Abstract pdf
- Earo Wang, Di Cook and Rob J Hyndman (2020) A new tidy data structure to support exploration and modeling of temporal data. Journal of Computational & Graphical Statistics, to appear. Abstract DOI pdf
- Pablo Montero-Manso, George Athanasopoulos, Rob J Hyndman, Thiyanga S Talagala (2020) FFORMA: Feature-based Forecast Model Averaging. International Journal of Forecasting, 36(1), 86-92. Abstract DOI pdf
Recent and upcoming seminars
- How Rmarkdown changed my life. (30 January 2020) More info...
- Tidy time series & forecasting in R. (27 January 2020) More info...
- The journal game. (29 October 2019) More info...
- Tidy time series analysis in R. (27 September 2019) More info...
- Feature-based time series analysis. (27 September 2019) More info...