November 21, 2019

What do statisticians do all day?

By

From the student dissertation talks this week

  • Toward Modeling of Disease Transmission Networks (E. coli in Germany)
  • Real Time Bus Headway Estimation in Auckland, New Zealand (the last bus was on time but your bus is late)
  • Survival Analysis of Guinea pigs with dental diseases (worse than you think)
  • Sample Path Behaviour of Accumulating Priority Queues
  • Comparison of the Bayesian and Simple Model for Estimations
  • Finding Periodic Climate Cycles in a Mud Core (we can’t see the sunspot cycle; we have a sad)
  • Optimal Sample Allocation for Estimating Regression Parameters (if it’s optimal for one purpose it won’t be for others)
  • Queue Mining — Online Delay Prediction
  • Enabling Text Analytics (simpler software)
  • Systematic Error Removal using Random Forests (in metabolomics)
  • Tracking of Dietary Patterns as Children Grow Up
  • Exploration of the Effects of a Text-MessageBased Diabetes Self-Management Support Programme (it works!)
  • A Study of Ticketing Prediction in the Events Industry (they didn’t give us the right data)
  • Anomaly Detection in Business Transactions UsingSupervised and— Unsupervised Methods (Fraud, we haz it)
  • Designing for a conceptual understanding of the Mean and Standard Deviation
  • Detecting Ecological Change along Environmental Gradients (for critters that live near the shore)
  • Identifying the Best Predictors for Power Demand Across Auckland
  • Automatic Identification of Patient Smoking Status based on Unstructured Clinical Notes (you’d think doctors would just say ‘smoker’. Sadly, no)
  • Visualization of Network Data (ooh, pretty)
  • Brownian Motions and Excursions
  • New methods for estimating population size based on close-kin genetics and extensions (Whales and inbreeding and population size)
  • An Examination of the Relationship between Student Engagement and Academic Achievement (it looks like lectures and tutorials are useful, but confounding)
avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »