Search
Search Funnelback University
- Refined by:
- Date: Past month
51 -
100 of
100
search results for news |u:www.statslab.cam.ac.uk
Fully-matching results
-
Sensitivity Analysis | Qingyuan Zhao
https://www.statslab.cam.ac.uk/~qz280/tag/sensitivity-analysis/3 Jun 2024: Search. Sensitivity Analysis. A new approach to sensitivity analysis in linear SEMs using stochastic optimization and the $R2$-calculus. -
Qingyuan Zhao | Qingyuan Zhao
https://www.statslab.cam.ac.uk/~qz280/author/qingyuan-zhao/3 Jun 2024: I am interested in improving the general quality and appraisal of statistical research, including new methodology and a better understanding of causal inference, novel study designs, sensitivity analysis, multiple testing, and -
IV & MR (project page) | Qingyuan Zhao
https://www.statslab.cam.ac.uk/~qz280/project/iv-mr/3 Jun 2024: A new Bayesian full-information method for life-course Mendelian randomization. Zixuan Wu, Ethan Lewis, Qingyuan Zhao, Jingshu Wang. ... The new GRAPPLE framework can handle multiple exposures and overlapping exposure and …. Jingshu Wang, Qingyuan Zhao, -
Exploratory Data Analysis | Qingyuan Zhao
https://www.statslab.cam.ac.uk/~qz280/tag/exploratory-data-analysis/3 Jun 2024: The increasing availability and scale of Genome Wide Association Studies (GWAS) bring new horizons for understanding biological mechanisms. -
Migrating my website to Hugo + Academic + ox-hugo | Qingyuan Zhao
https://www.statslab.cam.ac.uk/~qz280/post/migrating/3 Jun 2024: To add this new publication to my webpage, I can execute the following command. ... So my homepage only contains two widgets. Additionally, I created a custom widget page called news to make announcements and display new content. -
A note on post-treatment selection in studying racial discrimination…
https://www.statslab.cam.ac.uk/~qz280/publication/policing-risk-ratio/3 Jun 2024: We propose using an estimand new for this context—the causal risk ratio, which has more transparent interpretation and requires weaker identification assumptions. -
Lecture Notes on Causal Inference(with corrections) Qingyuan Zhao May …
www.statslab.cam.ac.uk/~qz280/teaching/causal-2023/notes-2021.pdf3 Jun 2024: Treatment Ai A = {0, 1}. Example: a new drug. – -
Statistical Modeling: Returning to its roots | Qingyuan Zhao
https://www.statslab.cam.ac.uk/~qz280/post/breiman-reprint-2021/3 Jun 2024: Scientific research is usually an iterative process. The cycle: conjecture-design-experiment-analysis leads to a new cycle of conjecture-design-experiment-analysis and so on. ... statisticians (including me) are trained to develop new techniques and most -
www.statslab.cam.ac.uk/~qz280/tag/philosophy/index.xml
www.statslab.cam.ac.uk/~qz280/tag/philosophy/index.xml3 Jun 2024: Philosophy | Qingyuan Zhao https://www.statslab.cam.ac.uk/qz280/tag/philosophy/ Philosophy Source Themes Academic (https://sourcethemes.com/academic/) en-us 2024 Tue, 10 Jan 2023 09:27:00 0000 -
Is this estimand really an average treatment effect? | Qingyuan Zhao
https://www.statslab.cam.ac.uk/~qz280/post/policing-ate-estimand/3 Jun 2024: The new paper shows that a weaker assumption called subset ignorability is sufficient for the identification. -
www.statslab.cam.ac.uk/~qz280/post/index.xml
www.statslab.cam.ac.uk/~qz280/post/index.xml3 Jun 2024: We invented some new notation to distinguish some different types of graph connections/separations. ... Macros for the new notation can be found in the file below:</p> <p> <a -
www.statslab.cam.ac.uk/~qz280/index.xml
www.statslab.cam.ac.uk/~qz280/index.xml3 Jun 2024: We invented some new notation to distinguish some different types of graph connections/separations. ... Macros for the new notation can be found in the file below:</p> <p> <a -
www.statslab.cam.ac.uk/~qz280/tag/causal-inference/index.xml
www.statslab.cam.ac.uk/~qz280/tag/causal-inference/index.xml3 Jun 2024: The cycle: conjecture-design-experiment-analysis leads to a new cycle of conjecture-design-experiment-analysis and so on. ... The new paper shows that a weaker assumption called subset ignorability is sufficient for the identification. -
SISCER Module 12 Causal Inference with Observational Data: Common…
www.statslab.cam.ac.uk/~qz280/teaching/siscer-2023/L0.pdf3 Jun 2024: I I develop new methods and apply causal inference methods to answer scientificquestions. ... I New methodology and better understanding of causal inference;. I Novel study designs;. -
Practical: Randomization inference
www.statslab.cam.ac.uk/~qz280/teaching/siscer-2023/S1.pdf3 Jun 2024: would be violated if, for example,the hospital staff need some time to adjust to the new system.). -
SISCER Module 12 Lecture 4: Matched Case-Control Studies
www.statslab.cam.ac.uk/~qz280/teaching/siscer-2023/L4.pdf3 Jun 2024: New England Journal of Medicine,284(15):878–881. Rosenbaum, P. R. (1991). Sensitivity analysis for matched case-control studies. -
STATISTICAL MODELLING Part IIC, Michaelmas 2021Practical 2: More on…
www.statslab.cam.ac.uk/~qz280/teaching/modelling-2022/P2.pdf3 Jun 2024: Writing functions. When writing anything but a short algorithm, it is often easiest to edit the commands from a text file.In Rstudio create a new Rscript using CtrlShiftN. -
SISCER Module 12 Lecture 1: Randomization inference
www.statslab.cam.ac.uk/~qz280/teaching/siscer-2023/L1.pdf3 Jun 2024: 12 wards in 2 hospitals were randomized to switch from an old model of weekendallied health services to no services, before adopting a new model of services.(You can visualize the ... Rosenbaum, P. R. (2002). Observational Studies. Springer Series in -
STATISTICAL MODELLING Part IIC, Michaelmas 2021Practical 6: Binomial…
www.statslab.cam.ac.uk/~qz280/teaching/modelling-2022/P6.pdf3 Jun 2024: new function, do? ... Fit a new model such that the z-test for the coefficient of mumMyopicanswers this question. -
STATISTICAL MODELLING Part IIC, Michaelmas 2021Practical 1:…
www.statslab.cam.ac.uk/~qz280/teaching/modelling-2022/P1.pdf3 Jun 2024: par(new=TRUE). Experiment with other distributions and other sample sizes. Exercises. -
STATISTICAL MODELLING Part IIC, Michaelmas 2021 Practical 8:…
www.statslab.cam.ac.uk/~qz280/teaching/modelling-2022/P8.pdf3 Jun 2024: One new piece of information in thesummary is the estimate of the dispersion parameter, which is based on the estimate. ... For a betteralternative, see the practical video.) Try fitting a new model with this term removed and store this inGammaMod0.Is -
PUBLICATIONS OF DOMINIC WELSH 1962 [1] J. M. Hammersley ...
www.statslab.cam.ac.uk/~grg/papers/welsh-bib.pdf6 Jun 2024: Springer, New York, 1969. [14] D. J. A. Welsh. Transversal theory and matroids. ... Math. Appl., volume 72, pages 133–148. Springer, New York, 1995. [80] L. -
STATISTICAL MODELLING Part IIC, Michaelmas 2021Practical 3: Linear…
www.statslab.cam.ac.uk/~qz280/teaching/modelling-2022/P3.pdf3 Jun 2024: Omitting the last observation pushes the estimatesfor the coefficients almost to the edge of their 23% confidence ellipsoid—not necessarily a big cause forconcern.Now create a new response y2 as ... predict.lm) can be used to give xT β̂ for a new -
STATISTICAL MODELLING Part IIC, Michaelmas 2019Practical 2: More on…
www.statslab.cam.ac.uk/~qz280/teaching/modelling-2022/P2S.pdf3 Jun 2024: Now, let’s see what happens if we make some of the true coefficients different from 0.# Compute t-statistics in 10000 simulations from the new modelt_stat_mat2 <- -
Confounder Selection via Iterative Graph Expansion Qingyuan Zhao…
www.statslab.cam.ac.uk/~qz280/talk/lse-2024/slides.pdf3 Jun 2024: They provided a new definition: “a pre-exposure covariate C for which there exists a set of othercovariates X such that effect of the exposure on the outcome is unconfounded conditional -
Using sparsity to overcome unmeasured confounding:Two examples…
www.statslab.cam.ac.uk/~qz280/talk/mrc-bsu-2019/slides.pdf3 Jun 2024: Old and new problems. Epidemiology and public health: effectiveness of prevention/treatment, causaleffect of risk factors, etc. ... Qingyuan Zhao (Stats Lab) Specificity MRC-BSU Seminar 27 / 30. New MR results. -
Small Data, Big Time—A retrospect of the first weeks of COVID-19
www.statslab.cam.ac.uk/~qz280/talk/rss-2021/slides.pdf3 Jun 2024: Jun 9 New naming system for variants of interest and concern. ... Two initial studies. 1 Q. Li, et al. Early transmission dynamics in Wuhan, China, of novel coronavirusinfected pneumonia.New England Journal of Medicine. -
Discovering Mechanistic Heterogeneity using Mendelian Randomization
www.statslab.cam.ac.uk/~qz280/talk/pcic-2020/slides.pdf3 Jun 2024: Genetic studies of body mass index yield new insightsfor obesity biology”. -
Confounder Selection via (Iterative) Graph Expansion Qingyuan Zhao…
www.statslab.cam.ac.uk/~qz280/talk/pcic-2023/slides.pdf3 Jun 2024: Outline. 1 Review. 2 New approach. Qingyuan Zhao (Cambridge) Confounder selection September 17, 2023 2 / 17. ... Outline. 1 Review. 2 New approach. Qingyuan Zhao (Cambridge) Confounder selection September 17, 2023 3 / 17. -
Multiple conditional randomization tests
www.statslab.cam.ac.uk/~qz280/talk/ims-2022/slides.pdf3 Jun 2024: How should we resolve this? Our proposal. Use a new term—quasi-randomization tests. ... 2 To provide a unifying framework that incorporates many old and new ideasabout multiple conditional randomization tests. -
Comment on “Causal inference using invariant prediction” Qingyuan…
www.statslab.cam.ac.uk/~qz280/publication/invariant-prediction-comment/comment.pdf3 Jun 2024: proteins. Using their own method, Sachs et al. [2005] reportedly recovered 15of the known directed arcs and discovered two new putative links (not shown),and missed 3 of the interactions which -
STATISTICAL MODELLING Part IIC, Michaelmas 2021Practical 4: More on…
www.statslab.cam.ac.uk/~qz280/teaching/modelling-2022/P4.pdf3 Jun 2024: Fit a new linear model omitting thisobservation and also omitting the Screens variable. -
Bootstrapping Sensitivity Analysis
www.statslab.cam.ac.uk/~qz280/publication/sensitivity-ipw/slides.pdf3 Jun 2024: Rosenbaum’s paradigm New bootstrap approach. Population Sample Super-population. Design Matching Weighting. ... Statistical Science, 17(3):309–321, 2002. P. R. Rosenbaum. Observational Studies. Springer New York, 2002. -
Simultaneous Hypothesis Testing using Internal Negative Controls…
www.statslab.cam.ac.uk/~qz280/publication/ranc/slides.pdf3 Jun 2024: Solid threshold: A “new” method. There are two ways to obtain this threshold:. -
Mendelian randomization
www.statslab.cam.ac.uk/~qz280/talk/mr-tutorial/slides.pdf3 Jun 2024: Advanced lipoprotein testingA “new” technology—Nuclear Magnetic Resonance (NMR)—can nowmeasure subclass traits such as. ... 39/40. MR screening for lipoprotein subclasses. Scientific TakeawayI The new results show clear heterogeneity among HDL. -
Selection bias in 2020 Qingyuan Zhao Statistical Laboratory,…
www.statslab.cam.ac.uk/~qz280/talk/ocis-2020/slides.pdf3 Jun 2024: Common mistake: New data Existing model = New results. Qingyuan Zhao (Stats Lab, Cambridge) Selection bias in 2020 Dec. -
Selective Inference for Effect Modification
www.statslab.cam.ac.uk/~qz280/publication/effect-modification/slides.pdf3 Jun 2024: An equivalent model is. Yi µy (Xi ) =(Ti µt (Xi ). ) (Xi ) i, i = 1,. , n. The new nuisance parameters µy (x) and µt (x) can bedirectly estimated from the data. -
The Cycle of Statistical Research Qingyuan Zhao Statistical…
www.statslab.cam.ac.uk/~qz280/talk/ccimi-2020/slides.pdf3 Jun 2024: About me. “New” University Lecturer in the Stats Lab. PhD (2011-2016) in Statistics from Stanford. ... New analysis for lipoprotein subfractionsIf there are different pathways, HDL subfractions may show heterogeneouseffects in Mendelian randomisation. -
What is a randomization test?
www.statslab.cam.ac.uk/~qz280/talk/toronto-2022/slides.pdf3 Jun 2024: 2 To provide a unifying framework that incorporates many old and new ideas in the literature. ... How should we resolve this? Our proposal. Introduce a new term—quasi-randomization tests. -
Design: The Elusive Principle of Statistics
www.statslab.cam.ac.uk/~qz280/talk/oxford-2024/slides.pdf3 Jun 2024: Rosenbaum, Paul R. (2010). Design of Observational Studies. Springer Series in Statistics.New York: Springer. ... A New Criterion for Confounder Selection”.In: Biometrics. Journal of the International Biometric Society 67.4, pp. -
SISCER Module 12 Lecture 3: Sensitivity analysis
www.statslab.cam.ac.uk/~qz280/teaching/siscer-2023/L3.pdf3 Jun 2024: The lazy statistician’s method- Sample with replacement to create a new sample of the same size as the study. ... Springer Series in Statistics. Springer, New York. Zhao, Q. (2018). On sensitivity value of pair-matched observational studies. -
MR Data Challenge 2019 — The role of lipoprotein subfractions in…
www.statslab.cam.ac.uk/~qz280/publication/mr-lipids/report.pdf3 Jun 2024: 2017. “Association Analyses Based on False Discovery Rate Implicate New Loci forCoronary Artery Disease.” Nature Genetics 49 (9): 1385. -
Two-Sample Instrumental Variable Analysis: Challenges and Some…
www.statslab.cam.ac.uk/~qz280/publication/two-sample-iv/slides.pdf3 Jun 2024: Part 2 New statistical methods for two-sample MR. Two-SampleIV. QingyuanZhao. Introduction. ... The two stage LS is not asymptotically efficient. Next:. Part 2 New statistical methods for two-sample MR using justsummary statistics. -
SISCER Module 12 Lecture 2: Matching for Cohort Studies
www.statslab.cam.ac.uk/~qz280/teaching/siscer-2023/L2.pdf3 Jun 2024: New England Journal of Medicine, 312(3):137–141. 29 / 29. Observational studies and causal inference Basics of matching Various issues and techniques in matching References. -
BETS: The dangers of selection bias in early analyses ...
www.statslab.cam.ac.uk/~qz280/talk/ido-2021/slides.pdf3 Jun 2024: The beginning of this project. On December 30, 2019, I saw rumours amout a new SARS-like virus on theinternet. -
Mendelian randomization: From genetic association to epidemiological…
www.statslab.cam.ac.uk/~qz280/publication/mr-partially-bayes/slides.pdf3 Jun 2024: I We will reassess the evidence for HDL using a new design andnew statistical methods of Mendelian randomization. ... 8/33. New methods for MRPart 1: Increased robustness to pleiotropyWe will derive an estimator that is robust to both. -
Towards Reliable Inference for Precision Medicine
www.statslab.cam.ac.uk/~qz280/talk/jsm-2021/slides.pdf3 Jun 2024: Anequivalent model is. Yi µy (Xi ) =(Ti µt (Xi ). ) (Xi ) i, i = 1,. ,n. I The new nuisance parameters µy (x) and µt (x) can be directly estimated from the ... J. D. Gibbons, I. Olkin, M. Sobel, Selecting and ordering populations: A new statistical -
Mendelian Randomization: Old and New Insights Qingyuan Zhao…
www.statslab.cam.ac.uk/~qz280/talk/epfl-2021/slides.pdf3 Jun 2024: Mendelian Randomization: Old and New Insights. Qingyuan Zhao. Statistical Laboratory, University of Cambridge. ... I New life of an old idea:. MR = Base causal inference on randomness in Mendelian inheritance. -
Selecting and ranking individualized treatment rules with…
www.statslab.cam.ac.uk/~qz280/talk/selective-inference-webinar-2021/slides.pdf3 Jun 2024: Many possible objectives for selection and ranking. Selective inference for partially identified/ordered problems: a potentially new topic? ... J. D. Gibbons, I. Olkin, M. Sobel, Selecting and ordering populations: A new statistical methodology, (SIAM, -
Will competition-winning methods for causal inference also succeed in …
www.statslab.cam.ac.uk/~qz280/publication/acic-competition-comment/paper.pdf3 Jun 2024: Rosenbaum, P. R. (2002). Observational Studies, 2nd ed. Springer, New York, NY.Rosenbaum, P. ... Design trumps analysis: Two new interpretations. The role of statistics in causal inference.
Refine your results
clear all
Date
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.