Michael Pyrcz, PhD, P.Eng (daytum Founding Advisor)
Statistics provide quantitative lens for a new perspective on subsurface data and to better understand subsurface uncertainty.
Learning Geostatistics will help you leverage:
- Deductive Statistics – Pooling data for the purpose of quantification of univariate, multivariate and spatial phenomenon
- Inferential Statistics – Methods to make inferences concerning the population from a sample. In the subsurface we only directly sample about 1 trillionth of the reservoir; therefore, essential!
- Frequentist Statistics – Drawing conclusions based on frequencies or proportions. So many problems can be solved by counting!
- Bayesian Statistics – Drawing conclusions based on updating belief with new information. Bayesian theorem can solved difficult problems such as the probability of a subsurface depositional setting, given well core data observations.
- Statistical Representativity – Sampling for representativity and treatment of bias. All subsurface datasets are biased!
- Statistical Significance – Testing the significance of results with hypothesis testing and confidence intervals. Does the result matter?
- Spatial Modeling – The subsurface has spatial structures; therefore, by capturing them we get much better models.
- Statistical Modeling – Data-driven modeling and prediction. Regression to machine learning, let’s join the data-driven paradigm!
- Uncertainty Models – Modeling, summarizing and making decisions in the presence of uncertainty. Uncertainty is ubiquitous and due to our ignorance.
- Big Data Analytics – Working with high volume, variety and velocity data inherent to subsurface exploration
When Michael is not building python packages or mentoring students, he’s either running, out on his Jeep, or kayaking around Lake Austin. You can find him on Twitter here, and his YouTube channel here.