Open Forecast

Discover R Packages for Forecasting

Slider
Forecasting with R
Discover R Packages for time series analysis and forecasting
Slider
Cutting edge research in forecasting
Forecasting the most difficult cases with open source software
Slider
Research in practice
Connect the dots between the modern research in forecasting and the applications
Banner Shadow

R Packages

Tutorials

About

Open Forecast is the website about forecasting packages for R, developed by the members of Centre for Marketing Analytics and Forecasting of Lancaster University, UK. The idea of this project is to make the cutting-edge research available to wide audience of forecasters (both practitioners and academics).

Why R?

R has already become a standard in the statistical and data science societies. It’s popularity is on the peak, and many businesses now consider using R instead of Excel and / or professional analytics systems for some of the tasks. The advantages of R can be summarised as:

  • Free, open source software (distributed under GPL license). This means that you don’t need to pay for R, you can just download, install and use it.
  • Free packages for R. This means that you have thousands of programs, models, modules and instruments for a wide variety of tasks. Are you interested in forecasting? There’s a dozen of packages on that topic. Are you thinking of doing some analysis with your data? Once again, you have a plethora of packages to choose from.
  • Flexible program. You don’t need to be a specialist in programming language in order to work with R. You don’t need to bother about declaring objects or setting their types. R is very smart and flexible to do those things for you.

However, with all the advantages, there are also several disadvantages:

  • Given that both R and R packages are free and open source, nobody is really responsible for a bad work of either program or a package in your company. If you encounter an issue, you need to find a way to resolve it on your own. Yes, the community is usually very helpful and package maintainers are responsive in bug fixes, but you might need to be slightly more proactive in finding and fixing the bugs, than you would be if you used some expensive proprietary software.
  • Flexibility has its costs. Given that you don’t need to declare any variables, means that R needs to spend some time, identifying them. This leads to the increase in the computational time. As a result, many R functions are slow, especially when it comes to applying them to large datasets. There are several solutions to this problem, some of which involve writing core functions either in C or C++. However, not many analysts or even academics are ready to learn additional languages to improve their code.

Given these advantages and disadvantages, every person should decide for himself, whether to go R route or not.

Open Forecast tries to smooth the edges and make R as accessible to wide audience as possible. In the development of R packages, we try to mitigate the disadvantages as much as possible.

Logo