**Venue:** The UnConference will take place at the University of Auckland 24–25 November, 2020. All sessions will be held in the Maths and Physics Building (Building 303), which
is located at 38 Princes Street. Rooms MLT2 and MLT3 are on the first floor, and SLT1 is on the ground floor. There is considerable construction going on around the campus, but the main doors on Princes Street will always be accessable.
This entry is at the basement level.

**Reception:** There is no formal reception, but some of us will gather at the Vultures' Lane in the CBD (map here) around 1800 on Monday evening.

**Session Chairs:** Session chairs for the keynote talks have been arranged. In all other sessions, the session chair will be the **last** speaker of that session. Speakers should aim to talk for 15-17 minutes, and allow for a couple of questions at most. Please try and stick to the
times so that people can move between rooms easily.

**Questions:** Remember that if you have a question for a speaker:

- It should be a question not a statement.
- It should be about the talk and not about you.
- It should be
**short**. - If you really think the speaker is wrong, then it might be better to talk to them when they are off-stage.
- To please be respectful.
- And finally, that James hates most questions, so think before you ask :-)

## Schedule

Time | Tuesday 24/11 | ||
---|---|---|---|

855 | Housekeeping | ||

900 | Education, democratizing data, and software: Targeting the intersection Chris Wild University of Auckland MLT2/303-102 | ||

950 | Morning Tea (30 minutes) | ||

MLT2/303-102 | MLT3/303-103 | SLT1/303-G01 | |

1030 | A Platform for Large-Scale Statistical Modelling using R: Preliminary ResultsJason CairnsUniversity of Auckland | Tree based credible set estimationKate LeeUniversity of Auckland | A framework to evaluate imputation strategies at Stats NZFelipa ZabalaStats NZ |

1050 | Constrained Maximum Likelihood for Correlated DataYu Jin KimUniversity of Auckland | A Continuous-time, discrete-space model of marine mammal exposure to Navy sonarCharlotte M. Jones-ToddUniversity of Auckland | A machine learning model to identify private dwellings from admin dataSusmita DasStats NZ |

1110 | Genealogies in branching populations: Many spines make light work...Simon HarrisUniversity of Auckland | Optimal sampling of generalized raking estimators for regression modelling in two-phase designsTong ChenUniversity of Auckland | Interactive Visualisation using RCloudSimon UrbanekUniversity of Auckland |

MLT2/303-102 | MLT3/303-103 | SLT1/303-G01 | |

1130 | Accessing evidence of firing pin impression by using machine learning.Jason WenUniversity of Auckland | The need for speed in Genomics: Comparing Bayesian algorithms to estimate polygenic effectsRoy CostillaUniversity of Queensland | |

1150 | Optimization of Inductive Linearisation – application to the Michalis–Menten modelSepi SharifUniversity of Otago | Modelling for COVID in Official Economic Time SeriesRichard PennyStats NZ | |

1210 | Lunch with AGM (1 hour 30 minutes) AGM in MLT2/303-102 | ||

1340 | A lifetime of data - Biometrics Technician to Senior Applied Statistician Maree Luckman Fonterra MLT2/303-102 | ||

MLT2/303-102 | MLT3/303-103 | SLT1/303-G01 | |

1430 | Designed experiments for tuning hyperparameters in machine learning algorithmsAgnes Yongshi DengUniversity of Auckland | There and back again: A statisticians journey into the `real world' and back to academia.Andrew BalemiUniversity of Auckland | Testing the confidentiality of synthetic data for the Stats NZ Integrated Data Infrastructure (IDI) Population Explorer datasetAlistair RamsdenStats NZ |

1450 | Online and alone: Designing positive first experiences with computer programming for statistics students learning remotelyAnna FergussonUniversity of Auckland | Using Bayesian Growth Models to Predict Grape YieldRory EllisFonterra | Data is a taonga: using data in a Aotearoa/New Zealand contextLinley JessonPlant and Food Research |

1510 | The Future of Statistics at New Zealand UniversitiesMartin HazeltonUniversity of Otago | The Implementation of Biological Models for the Probabilistic Interpretation of NGS aSTR MixturesKevin ChengUniversity of Auckland | clustglm and clustord: 2 R packages for model-based clustering of binary, count and ordinal data with covariatesLouise McMillanVictoria University of Wellington |

1530 | Afternoon tea (20 minutes) | ||

MLT2/303-102 | MLT3/303-103 | SLT1/303-G01 | |

1550 | Sounds like RandomnessAmy RenelleUniversity of Auckland | Reproducible Research with DockerGlenn ThomasHarmonic Analytics | HLFS mode of collection: A journey due to COVID-19Wilma MolanoStats NZ |

1610 | Two-phase subsampling design for DNA sequencing with application in the relatedness in endangered speciesPei LuoUniversity of Auckland | Improving the prediction of bus arrival using real-time network stateTom ElliottVictoria University of Wellington | Non-negative forecast reconciliation for forecasting hierarchical time seriesShanika WickramasuriyaUniversity of Auckland |

1630 | Optimal sampling allocation for outcome dependent designs in cluster-correlated data settingsClaudia Rivera-RodriguezUniversity of Auckland | Practical Assessment of Spatial Capture-RecaptureDavid ChanUniversity of Auckland | Overcoming singularity: a Khmaladze transform goodness of fit test for the Laplace distributionJohn HaywoodVictoria University of Wellington |

1650 | Estimating the time lag between predator abundance and prey abundanceMartin UpsdellAgResearch | simGBS: Unlimited Genotyping-by-Sequencing Data for FreeJie KangUniversity of Otago | Integrating R Graphics and TikZ GraphicsPaul MurrellUniversity of Auckland |

1830 | Conference Dinner Venue: Old Government House | ||

1700 |

Time | Wednesday 25/11 | ||
---|---|---|---|

855 | Housekeeping | ||

900 | Statistics of Ambiguous Rotations Richard Arnold VUW MLT2/303-102 | ||

950 | Morning Tea (30 minutes) | ||

MLT2/303-102 | MLT3/303-103 | SLT1/303-G01 | |

1030 | Missing in action - a statistical window on prisonsLen CookIGPS VUW | My Journey to create shiny app ‘DeltaGen’Dongwen LuoAgResearch | War StoriesPeter MullinsUniversity of Auckland |

1050 | Influence functions, and why you should careThomas LumleyUniversity of Auckland | Accuracy of the saddlepoint approximation for MLEsJesse GoodmanUniversity of Auckland | Dimension reduction for imbedding high dimensional measurements into Bayesian NetworksBeatrix JonesUniversity of Auckland |

1110 | A Bayesian approach to modelling of Phosphorus inputs to rivers from diffuse and point sources.Alasdair NobleAgResearch | Working with UNITAR to design an e-learning course for measuring progress on the UN's Sustainable Development IndicatorsJohn Harraway and Sharleen ForbesUniversity of Otago | Using canonical correspondence analysis and redundancy analysis to fit nonlinear gradients to community dataRussell MillarUniversity of Auckland |

MLT2/303-102 | MLT3/303-103 | SLT1/303-G01 | |

1130 | Keeping Things Running Smoothly: A Collection of Kernel-based CollaborationsTilman DaviesUniversity of Otago | Beyond the Integrated Data Infrastructure - building a strategic data resource for AotearoaAndrew SporleUniversity of Auckland | A versatile discrete distributionRolf TurnerUniversity of Auckland |

1150 | Decision Making for Partially Observable Markov ProcessesAzam AsanjaraniUniversity of Auckland | Adversarial Risk Analysis for modelling strategic adversaryChaitanya JoshiUniversity of Waikato | |

1210 | Closing Ceremony | ||

1230 | Lunch (1 hour 10 minutes) | ||

1340 | Conference Finish |