Software we work in, and what each is good for
| Software | Where it shines | Typical tasks we help with |
|---|---|---|
| SAS | Clinical trials, epidemiology, large administrative data | DATA step cleaning, PROC MEANS, FREQ, TTEST, GLM, MIXED, LOGISTIC, PHREG, FACTOR |
| EViews | Econometrics and time series | Unit root tests, ARDL and bounds testing, VAR and VECM, cointegration, GARCH, panel models |
| Python | Flexible data work and machine learning | pandas, statsmodels, scipy, scikit-learn, matplotlib and seaborn |
| MATLAB | Engineering, signal processing, simulation | Statistics and Machine Learning Toolbox, custom estimation, Monte Carlo |
| Minitab | Quality improvement and teaching | Control charts, capability analysis, design of experiments, regression |
| jamovi and JASP | Free, point-and-click analysis | t-tests, ANOVA, regression, factor analysis, Bayesian alternatives |
SPSS and R have dedicated pages of their own: see our SPSS help and R and RStudio help. If you are not tied to any package, we will suggest the one that best fits your method, your budget and what your department supports.
A closer look at SAS
SAS remains common in pharmaceutical research and public health because its procedures are well validated and its logs make every step auditable. It can be intimidating at first, mainly because the syntax separates data preparation (the DATA step) from analysis (PROC steps).
Two questions come up constantly. First, where to find the p-value: each procedure prints it in its own table, labelled Pr > |t|, Pr > F or Pr > ChiSq depending on the test. Second, how to run a paired t-test: PROC TTEST with a PAIRED before*after; statement compares two measurements on the same subjects. To save results as a PDF, wrap the procedures between ODS PDF FILE="results.pdf"; and ODS PDF CLOSE;. SAS OnDemand for Academics gives students free browser-based access, and we can work in that environment with you.
Econometrics in EViews
EViews makes time-series work quick, which also makes it easy to skip steps. We help you check stationarity with ADF, Phillips-Perron or KPSS tests before choosing between models, select lag length with information criteria, and run the diagnostics examiners look for: serial correlation, heteroskedasticity, normality of residuals and stability tests such as CUSUM. If your data are panels rather than single series, we can compare fixed and random effects with a Hausman test, or move the work into Stata if your supervisor prefers. Interpreting results matters as much as running them: a significant cointegrating relationship, for instance, says the series move together in the long run, not that one causes the other. Our regression analysis guide covers the foundations.
How a request usually unfolds
You send the data, your research question, the software and version you have access to, and any code or output so far. We confirm the analysis plan in writing, then build the program or workflow, keeping a record of each choice. You receive the files with annotations and, if you wish, a screen-sharing session to run everything yourself. Smaller fixes often turn around within one or two days; multi-model projects take longer. A free quote is the quickest way to get a realistic estimate.
Why explanation comes with every file
Output from any package is only useful if you can say what it means and why it was produced that way. Examiners and reviewers rarely care which software you used, but they will ask why you chose a model and whether its assumptions hold. Our annotated files and methods notes prepare you for those questions. For theses, you remain the analyst and should follow your institution's guidance on assistance, as described in our academic integrity approach.
What you get
- Commented program files: SAS .sas programs, EViews workfiles and programs, Python scripts or notebooks
- Correctly specified models with assumption checks appropriate to the method
- Annotated output explaining estimates, standard errors, confidence intervals and p-values
- Tables and figures formatted for your thesis or manuscript
- Exported results in the format you need, such as PDF, RTF or Excel
- A short methods note describing the procedure in software-neutral language
What we don't do
- Completing graded programming assignments or exams; for coursework we tutor you through your own attempts
- Supplying or cracking software licences; we work with versions you can legally access, including free alternatives
- Enterprise SAS administration or server deployment, which falls outside academic support
How it works
Describe your project
Level, deadline and what you need help with.
Agree a written scope
What's included, the price and the timeline.
Work in stages
Check-ins so you can steer the work.
Review and ask
Explanations and follow-up questions included.