Peter Austin

Statistician working in statistical methods for observational studies. Interested in propensity score methods for causal inference. All views my own.

[Peter Austin @Stats_research]

Graphical methods to illustrate the nature of the relation between a continuous variable and the outcome when using restricted cubic splines with a Cox proportional hazards model. Open access paper in SMMR. #statstwitter #stats https://t.co/Yk7aQE9cEC

[Peter Austin @Stats_research]

New open access paper on Imputation of incomplete ordinal and nominal data by predictive mean matching in Statistical Methods for Medical Research. https://t.co/ypTreHqxD1

[Peter Austin @Stats_research]

Open access paper on evaluating the median p-value method for assessing the statistical significance of tests when using multiple imputation https://t.co/MUXgQhuCEs

[Peter Austin @Stats_research]

Open access paper on a Variance Estimator for Marginal Cox Regression Models Fit to Non-Nested Multilevel or Clustered Data. #multilevel #statistics #clustereddata #survivalanalysis https://t.co/uORDD9M4Zd

[Peter Austin @Stats_research]

I’ll be posting on Bluesky going forward:
@stats-research.bsky.social

[Peter Austin @Stats_research]

New open-access paper: Inverse Probability of Treatment Weighting Using the Propensity Score With Competing Risks in Survival Analysis. #stats #statstwitter https://t.co/0N7WmBr9Eg

[Peter Austin @Stats_research]

The relative data hungriness of unpenalized and penalized logistic regression and ensemble-based machine learning methods: the case of calibration | Diagnostic and Prognostic Research. New open access paper. #stats #statstwitter. https://t.co/gsasHZEBEB

[Peter Austin @Stats_research]

A Comparison of Variance Estimators for Logistic Regression Models Estimated Using Generalized Estimating Equations (GEE) in the Context of Observational Health Services Research. New open access paper. #stats #statstwitter https://t.co/Dz5vOhZVaC

[Peter Austin @Stats_research]

Evaluating the median p-value method for assessing the statistical significance of tests when using multiple imputation. New open access paper published in Journal of Applied Statistics. #stats #statstwitter https://t.co/MUXgQhuCEs

[Peter Austin @Stats_research]

Using Multilevel Models and Generalized Estimating Equation Models to Account for Clustering in Neurology Clinical Research | Neurology. Open access tutorial article on statistical methods for the analysis of multilevel and clustered data. https://t.co/7dCQaF3f5h

[Peter Austin @Stats_research]

New open access paper on multiple imputation with competing risk outcomes using complex clinically realistic simulations: https://t.co/3ilxrqPdt3

[Peter Austin @Stats_research]

New paper on the effect of number of clusters and magnitude of within‐cluster homogeneity in outcomes on the performance of four variance estimators for a marginal multivariable Cox regression model fit to clustered data in observational research https://t.co/ftrYgsURaF

[Peter Austin @Stats_research]

Just published in JAMA: an introductory tutorial: Accounting for Competing Risks in Clinical Research with ⁦@hein_putter⁩ and @mar1a_ib. https://t.co/uxGEHcIigK

[Peter Austin @Stats_research]

New open access paper in Statistical Methods in Medical Research on the performance of marginal structural models (MSMs) for estimating risk differences and relative risks using weighted GLMs: https://t.co/1mvGmFEqfU. #propensityscore #statstwitter

[Peter Austin @Stats_research]

New open access paper in SMMR comparing predictive mean matching (PMM) with logistic regression for imputing missing binary variables when using multiple imputation: https://t.co/rZWYZ8dYOh. PMM had comparable performance, but required less processing time.

[Peter Austin @Stats_research]

New open access paper comparing target estimands for five sets of propensity score-based weights: ATE weights, ATT weights, overlap weights, matching weights, and entropy weights: https://t.co/48o3UXPzMR #statstwitter

[Peter Austin @Stats_research]

New paper in Statistics in Medicine: how to impute variables that are used in inclusion/exclusion criteria and that are independent variables in the analysis model of scientific interest: https://t.co/HHln5K5pBj with @DGiardiello_ and Stef van Buuren. #statstwitter

[Peter Austin @Stats_research]

The iterative bisection procedure: newly-published open access paper describing a useful tool to determine parameter values for Monte Carlo simulations: https://t.co/KJjfBkFvrg

#Statstwitter

[Peter Austin @Stats_research]

New open access paper with @ESteyerberg, @f2harrell
, @DouglasLeeMD using empirical analyses and simulations to compare machine learning with statistical models for predicting discharge blood pressure in patients hospitalized with AMI or CHF: https://t.co/pCnnc3tR1o

[Peter Austin @Stats_research]

New paper on assessing the calibration of competing risk regression models: https://t.co/lLJEW9uiJg with @hein_putter , Daniele Giardiello, and David van Klaveren.