Hastie, Tibshirani and Friedman’s Elements of Statistical Learning first appeared in 2001 and is already a classic. It is my go-to book when I need a quick refresher on a machine learning algorithm. I like it because it is written using the language and perspective of statistics, and provides a ... [Read more...]

There is no shortage of time series data available on the web for use in student projects, or self-learning, or to test out new forecasting algorithms. It is now relatively easy to access these data sets directly in R. M Competition data The 1001 series from the M-competition and the 3003 series ... [Read more...]

Last week, version 5.0 of the forecast package for R was released. There are a few new functions and changes made to the package, which is why I increased the version number to 5.0. Thanks to Earo Wang for helping with this new version. Handling missing values and outliers Data cleaning is ... [Read more...]

It is common to use a Ljung-Box test to check that the residuals from a time series model resemble white noise. However, there is very little practical advice around about how to choose the number of lags for the test. The Ljung-Box test was proposed by Ljung and Box (Biometrika, 1978) ... [Read more...]

Last year I taught an online course on forecasting using R. The slides and exercise sheets are now available at www.otexts.org/fpp/resources/ [Read more...]

We are looking for a new post-doctoral research fellow to work on the project “Macroeconomic Forecasting in a Big Data World”.  Details are given at the link below jobs.monash.edu.au/jobDetails.asp?sJobIDs=519824 This is a two year position, funded by the Australian Research Council, and working with ... [Read more...]

There are quite a few R packages available for nonlinear time series analysis, but sometimes you need to code your own models. Here is a simple example to show how it can be done. The model is a first order threshold autoregression:     where is a Gaussian white noise series with ... [Read more...]

The Centre for Forecasting at Lancaster University is conducting some research on judgmental forecasting and model selection. They hope to compare the performance of judgmental model selection with statistical model selection, in order to learn how to best design forecasting support systems. They would like forecasting students, practitioners and researchers ... [Read more...]

We are currently advertising for three academic positions, suitable for recent PhD graduates. Lecturer (Applied Statistics or Operations Research) Five-year position with MAXIMA and the School of Mathematical Sciences Two positions available. Applications close 31 October. More information. Lecturer (Econometrics/Business Statistics) Continuing position with the Department of Econometrics and Business ... [Read more...]

I’ve been getting emails asking questions about my upcoming course on Forecasting using R. Here are some answers. Do I need to use the Revolution Enterprise version of R, or can I use open-source R? Open source R is fine. Revolution Analytics is organizing the course, but there is ... [Read more...]

The following video has been produced to advertise my upcoming course on Forecasting with R, run in partnership with Revolution Analytics. The course will run from 21 October to 4 December, for two hours each week. More details are available at http:/... [Read more...]

I’ve had several emails recently asking how to forecast daily data in R. Unless the time series is very long, the simplest approach is to simply set the frequency attribute to 7. y [Read more...]

I am teaming up with Revolution Analytics to teach an online course on forecasting with R. Topics to be covered include seasonality and trends, exponential smoothing, ARIMA modelling, dynamic regression and state space models, as well as forecast accuracy methods and forecast evaluation techniques such as cross-validation. I will talk ... [Read more...]

This week I’ve been at the R Users conference in Albacete, Spain. These conferences are a little unusual in that they are not really about research, unlike most conferences I attend. They provide a place for people to discuss and exchange ideas on how R can be used. Here ... [Read more...]

Akaike’s Information Criterion (AIC) is a very useful model selection tool, but it is not as well understood as it should be. I frequently read papers, or hear talks, which demonstrate misunderstandings or misuse of this important tool. The following points should clarify some aspects of the AIC, and ... [Read more...]

This question was posed on crossvalidated.com: I have a monthly time series (for 2009–2012 non-stationary, with seasonality). I can use ARIMA (or ETS) to obtain point and interval forecasts for each month of 2013, but I am interested in forecasting the total for the whole year, including prediction intervals. Is there ... [Read more...]

My new online forecasting book (written with George Athanasopoulos) is now completed. I previously described it on this blog nearly a year ago. In reality, an online book is never complete, and we plan to continually update it. But it is now at the point where it is suitable for ... [Read more...]

The ETS modelling framework developed in my 2002 IJF paper (with Koehler, Snyder and Grose), and in my 2008 Springer book (with Koehler, Ord and Snyder), is now available in EViews 8. I had no idea they were even working on it, so it was quite a surprise to be told that EViews ... [Read more...]

When I want to insert figures generated in R into a LaTeX document, it looks better if I first remove the white space around the figure. Unfortunately, R does not make this easy as the graphs are generated to look good on a screen, not in a document. There are ... [Read more...]

It is common to fit a model using training data, and then to evaluate its performance on a test data set. When the data are time series, it is useful to compute one-step forecasts on the test data. For some reason, this is much more commonly done by people trained ... [Read more...]