{"id":25441381,"url":"https://github.com/bertcarnell/jsm2024","last_synced_at":"2025-05-15T17:12:23.730Z","repository":{"id":252549457,"uuid":"840763558","full_name":"bertcarnell/jsm2024","owner":"bertcarnell","description":null,"archived":false,"fork":false,"pushed_at":"2024-08-11T20:56:47.000Z","size":68,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2024-08-12T17:39:26.290Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bertcarnell.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-08-10T15:59:35.000Z","updated_at":"2024-08-11T20:56:50.000Z","dependencies_parsed_at":"2024-08-10T17:25:31.727Z","dependency_job_id":null,"html_url":"https://github.com/bertcarnell/jsm2024","commit_stats":null,"previous_names":["bertcarnell/jsm2024"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bertcarnell%2Fjsm2024","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bertcarnell%2Fjsm2024/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bertcarnell%2Fjsm2024/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bertcarnell%2Fjsm2024/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bertcarnell","download_url":"https://codeload.github.com/bertcarnell/jsm2024/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":239288758,"owners_count":19614180,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-02-17T12:27:58.376Z","updated_at":"2025-02-17T12:27:58.925Z","avatar_url":"https://github.com/bertcarnell.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# jsm2024\n\nThe Joint Statistical Meetings (JSM) were held in Portland, Oregon Aug 3-8, 2024.\n\n![JSM Logo](img/jsm_2024_logo.png)\n\n## NISS Luncheon (8/4 1200)\n\n### Veridical Data Science\n\nTalked with Professor Bin Yu (UC Berkeley) about her book: [Veridical Data Science](https://vdsbook.com/).\n\n### AI in Practice:  Transitions and Tools\n\nDr. Nancy McMillan, Battelle\n\n### Harvard Data Science Review\n\nXiao-Li Meng showed this great resource [here](https://hdsr.mitpress.mit.edu/) and [here](https://datascience.harvard.edu/)\n\n## Recent Developments in Causal Inference (8/4 1600)\n\n### A Meta-Learning Method fo Estimation of Causal Excursion Effects to Assess Time-Varying Moderation\n\nMicro-randomized Trials - The idea that you can treat a longitudinal study where the same treatment is applied multiple times and treat it like a raondomized trial.\n\nApplications to Next Best Offer are possible.\n\nJieru Shi - University of Michigan\n\nNeed to learn more about doubly robust estimators for causal inference:  They estimate the Averate Treatment Effect using both regression and inverse probability weighting methods.\n\n### Other Presentation\n\nBeeswarm plots [here](https://r-graph-gallery.com/beeswarm.html) and [here](https://shap.readthedocs.io/en/latest/example_notebooks/api_examples/plots/beeswarm.html)\n\nCan we add this to the torndado package?\n\n## 2024 ASA Statistics in Marketing Doctoral Research Award Finalists Presentation (8/4 0830)\n\n### A Scalable Recommendation Engine for New Users and Items - Boya Xu - Duke\n\n### Optimal Comprehensible Targeting\n\nMethod to make decision tree-type comprehensible targeting narratives.  - Walter Zhang\n\nPrice Elasticity\n\n### Real Time Personalization in Dynamic Environments - Hong Deng\n\n### Selecting Data and Parameter Granularities:  A Bayesian Dual-Network Clustering Approach - Mingyung Kim - Wharton School\n\n### Targeted Marketing with Large Batches - Keyan Li - MIT\n\n\n\n## Bayesian Causal Inference (8/5 1030)\n\n### A Semiparametric Bayesian Approach for Extreme Quantile Treatment Effects for Heavy-tailed data - Arnab  Aich\n\nCould be useful for balances\n\n### Bayesian Causal Prediction:  Multivariate Graphical Dynamic Models - Luke Vrotsos - Duke\n\nGreat talk on how to do Bayesian Synthetic controls.\n\nWith Mike West at Duke\n\nreferences:\n\n- [Comparative Politics and the Synthetic Control](https://economics.mit.edu/sites/default/files/publications/Comparative%20Politics%20and%20the%20Synthetic%20Control.pdf)\n- [The Economic Costs of Conflict: A Case Study of the Basque Country](https://economics.mit.edu/sites/default/files/publications/The%20Economic%20Costs%20of%20Conflict.pdf)\n- [Inferring causal impact using Bayesian structural time-series models](https://arxiv.org/abs/1506.00356)\n- [GPU Accelerated Bayesian Learning and Forcasting in Simultaneous Graphical Dynamic Linear Models](https://projecteuclid.org/journals/bayesian-analysis/volume-11/issue-1/GPU-Accelerated-Bayesian-Learning-and-Forecasting-in-Simultaneous-Graphical-Dynamic/10.1214/15-BA946.pdf)\n    - [code](https://github.com/lutzgruber/gpuSGDLM)\n\n### Bayesian Sensitivity Analysis for Extending Inferences to a Target Population without Positivity - Jun Lu - University Illinois\n\n### Causally Sound Priors for Binary Experiments - Nicholas Irons - U Washington\n\n### Horeshoe Priors for Sparse Dirichlet-Multinomial Models - Yuexi Want - University of Illinois\n\n### Incorporate external control data into RCT using propensity score stratification and mixture prior - Xun Xu - U Texas\n\n\n\n## Advanced Statistical Methods in Non-parametric Statistics and Causal Inference for Complex Data Structures (8/5 1400)\n\n### Identifying Significant Mediators in High Dimensional Causal Graphical Models with FDR Control - Xiufan Yu, Notre Dame\n\n### De-confounding Causal Inference using Latent Multiple-Mediator Pathways - Yubai Yuan - Penn State\n\n### Causal Inference on Distribution Functions - Dehan Kong - U Toronto\n\n[paper](https://arxiv.org/abs/2101.01599)\n[code](https://github.com/kongdehanstat/causaldistributionfunction)\n\n### Nonlinear Global Frechet Regression for Random Objects via Weak Conditional Expectation - Satarupa Bhattacharjee - Penn State\n\n### Smoothed Robust Phase Retrieval - Zhong Zheng - Penn State\n\n\n## The Application of Generative Models in Marketing Research and Practice (8/6 1030)\n\n### Using GPT for Market Research - Ayelet Israeli - Harvard Business School\n\n- Used GPT as a random customer sampling mechanism and examined the distribution of its responses\n- When prompted as a random customer, GPT exhibits behaviors consistent with economic theory (price elasticity)\n- GPT-based estimates are realistic and consistent with values obtained from existing research\n\n### Using Multimodal LLM to Extract and Discover Features from Ad Images - Poppy Zhang, Meta\n\nTrained a classifier to extrat the important parts of an image for Meta customers\n\n[nielsen study](https://www.nielsen.com/insights/2017/when-it-comes-to-advertising-effectiveness-what-is-key/)\n\nLots of other Insights from Nielsen\n - [here](https://www.nielsen.com/insights/)\n - [here](https://www.nielsen.com/insights/2022/roi-report/)\n - [here](https://www.nielsen.com/insights/2024/are-you-investing-performance-marketing-for-right-reasons/)\n\nMulti-model LLM\n\n[LLAVA](https://llava-vl.github.io/)\n\nResults:  Top Advertisers (controlling for sales region, industry, and campaign size)\n\n- show price (negative)\n- contain promotion\n- show positive emotion\n- show experience of using the product\n- comparative\n- show endorsement\n- use humor\n- visually complext (negative)\n\n## IMS Medallion Lecture:  Winners with Confidence:  Discrete Argmin Inference Using Cross-Validated Exponential Mechanism - Ji Zhu (8/6 1400)\n\n[paper](https://arxiv.org/abs/2408.02060)\n\nIdea was how to say which machine learning results were actually significantly different from each other\n\n## Robust Causal Inference (8/7 0830)\n\n### Adaptive Experiments with Delayed Outcomes - Waverly Wei - USC\n\nCreate adaptive experiments where you measure some things early, and then other measures later, and you want to adaptively change the experiment\n\n### Bias Correction for Randomization-Based Average Treatment Effect Estimation in Inexactly Matched Observational Studies - Siyu Heng - NYU\n\n### Design-based causal inference for balanced incomplete block designs - Nicole Pashley - Rutgers\n\n[Nicole Pashley](https://sites.google.com/view/npashley/home?authuser=0)\n[Taehyeon Koo](https://taehyeonkoo.github.io/)\n\n### Random distribution shift and causal inference - Zhonghua Liu - Columbia\n\n\n\n## New Methods in Causal Inference and Reinforcement Learning for Personalized Decision-Making (8/7 1030)\n\n### The promises of multiple outcomes - Linbo Wang - U Toronto\n\n[website](https://sites.google.com/site/linbowangpku/home)\n[paper](https://arxiv.org/abs/2012.05849)\n[slides](https://drive.google.com/file/d/1jwkefODWm7KBSzl0RNMUT9MABIC8YbIE/view)\n\n### Balancing Personalization and Pooling: Decision-making and Statistical Inference with Limited Time Horizons - Yongyi Guo\n\n### Did We Personalize? Assessing Personalization by an Online Reinforcement Learning Algorithm Using Resampling - Raaz Dwivendi - UC Berkeley\n\n### Further Results on Target Trials and Structural Nested Models: Emulating RCTs using Observational Longitudinal Data - Anish Agarwal - Columbia\n\n\n\n## Causal Inference on Networks and Complex Data Structures (8/7 1400)\n\n### Analyzing Political Polarization in the Presence of Partially Observed Social Networks - Sharmodeep Bhattacharyya, Oregon State University\n\n### Causal Inference Under Network Interference with Noise - Daniel Sussman, Boston University\n\n### Exploratory Data Analysis, Confirmatory Data Analysis and Replication in the Same Observational Study: A Two Team Cross-Screening Approach to Studying the Effect of Unwanted Pregnancy on Mothers' Later Life Outcomes - Dylan Small, University of Pennsylvania\n\n### Nonsense Associations in Markov Random Fields - Elizabeth Ogburn, Johns Hopkins University\n\n\n## COPSS Distinguished Achievement Award and Lectureship (8/7 1600)\n\n#### Pre-Training and the Lasso\n\n[paper](https://arxiv.org/abs/2401.12911)\n[paper](https://arxiv.org/html/2401.12911v1)\n[similar presentation - same slides](https://www.youtube.com/watch?v=zIGc5Z2MaYM)\n[ptLasso](https://github.com/erincr/ptLasso/)\n[web](https://erincr.github.io/ptLasso/)\n[Erin Craig](https://www.erincraig.me/)\n\n\n\n\n\n\n\n\n\n\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbertcarnell%2Fjsm2024","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbertcarnell%2Fjsm2024","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbertcarnell%2Fjsm2024/lists"}