Estimation of the Fraction of Matter and Energy of the Universe
Situation: What is the fractional amount of matter and energy of the universe? This project aim was to answer that question, comparing mathematical models and a Machine Learning algorithm to observational mesuares.
Task: Analyse luminosity distance data to estimate the cosmological parameters using statiscs and advanced algorithms.
Action: I collected supernova data from an international collaboration survey, built from scratch a non-naive bayes regression algorithm to fit the luminosity distance curve and used Bayesian inference to find the probabilities of the parameters values, given the data collected. Then, I used the Markov Chain Monte Carlo method to estimate the final values and respective uncertainties.
Result: The fractions of matter and radiation that I estimated are in agreement with the Planck Satellite 2018 release. These parameters was computed considering 10000 Markov Chains generated by 2000 walkers using the emcee implementation of the Monte Carlo algorithm. Since about 40 steps are needed for the chain to “forget” where it started we discarded the initial 100 steps. The final estimation choosen was the 50-th percentile of the flatted samples and the uncertainty the diference of the central value with the 25-th and the 75-th percentiles respectively. The final values encountered are matter density
which agree with the direct observation of these parameters released by th Planck Satelite collaboration (2018):
See the complete project repository.
More Projects
- Denominational School Analysis
- Estimation of the Fraction of Matter and Energy of the Universe
- Space X Falcon 9 Landing Success Prediction
- YouTube Channel Evolution During Pandemics
Contact
Feel free to reach out to me on LinkedIn or contact by email dimas.jackson.ds@gmail.com. I’m always open to interesting discussions and collaborations!
Thanks for visiting!