Statistical analysis in cosmology
These lecture series will involve familiarizing the participants with statistical tools that are often employed in cosmological analysis for:
- Modelling of data and statistical inference
- Goodness of fit and confidence intervals
- Marginalisation and combining different experiments
- Determining and quantifying possible agreements or disagreements
- Introduction to machine learning
Analysis of observational datasets
The summer school will aim to train the participants in analysing big data sets from present and future large-scale surveys. This training will include:
- How to download and preprocess galaxy imaging data from the HST archive.
- How to define a likelihood function to this data, for a specific science case.
- How to fit a model to the data, via this likelihood function, using Bayesian inference.
- Performing Bayesian hierarchical analysis of a large imaging dataset.
Hands-on cosmological N-body simulations
N-body simulations enable us to compare cosmological models and theories with ever-more powerful data sets obtained from large telescopes and cosmic probes. In this domain, we will organize lectures that would cover:
- The theory and principles underlying N-body simulations.
- Setting up N-body simulations: This would include hands-on some publicly available simulation codes (like Gadget, COLA, SWIFT).
- Running simulations and post-processing the simulation data for specific scientific purposes.
- Reading and analysing the simulation data.
