Forget about Excel. It is hopeless for any serious work. (If you need convincing, see Errors, faults and fixes for Excel.) Any research student in the quantitative sciences should be using a matrix language for computation. I recommend students use R. It is free, has a big user base, and has zillions of add-on packages.
Blog posts about R
- Animated plots in R and LaTeX
- Blog aggregators
- Building R packages for Windows
- Comparing HoltWinters() and ets()
- CrossValidated Journal Club
- CrossValidated launched!
- Data visualization
- Data visualization videos
- Debugging in R
- Different results from different software
- Econometrics and R
- Exponential smoothing and regressors
- Finding an R function
- Forecast estimation, evaluation and transformation
- Forecasting time series using R
- Forecasting with long seasonal periods
- Forecasting workshop: Switzerland, June 2011
- Forecasts and ggplot
- Happy World Statistics Day!
- How to avoid annoying a referee
- Initializing the Holt-Winters method
- Installing R
- Internet surveys
- Kaggle on TV
- Learning R by video
- Major changes to the forecast package
- Measuring time series characteristics
- More StackExchange sites
- Organization and R
- R books
- R graph with two y-axes
- R help links
- R help on StackOverflow
- R workshop
- Research position in forecasting renewable energy
- RStudio: just what I’ve been looking for
- Six places left for the forecasting workshop
- Stack exchange for statistical analysis needs you!
- Statistical Analysis StackExchange site now available
- Statistical tests for variable selection
- The ARIMAX model muddle
- The art of R programming
- Time series cross-validation: an R example
- Time series packages on R
- Twenty rules for good graphics
- Using Google Reader
- What should we call the stats Q&A site?
- What you wish you knew before you started a PhD
- Why R is better than Excel for teaching statistics
- Workflow in R

Rob J Hyndman