Institute of Statistical Science Academia Sinica [Seminar Feed] http://www.stat.sinica.edu.tw Statistics, Stat, Edu en-us Tue, 11 Dec 2018 03:12:34 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/rss/ PHP admin@stat.sinica.edu.tw admin@stat.sinica.edu.tw Clinical Trial and the Working Experience in CRO http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2317/ Abstract

   

    In this talk, I will give a brief introduction to Clinical Trial, including the four main phases of the drug development process, the current situation of the drug industry, and the challenges we will confront. In the second part, I will introduce the basic structure of a global CRO company -- how people with different expertise working together -- and share my working experience as Clinical Data Analyst in PAREXEL for 2 years.

]]>
Fri, 30 Nov 2018 13:10:20 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2317/
智慧製造應用探討-大數據分析與機器學習的產業實例 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2318/ Abstract

   

    IC產業為貢獻台灣大量產值的重要命脈,不論是在IC製造或封裝上都是世界第一的供應者。本次的主題將介紹基本封裝之流程與挑戰以及機器學習在產業界的應用實例—傳統的IC封裝過程中往往需要依靠工程師長年累積的經驗,才能有效率的完成配線設計,下線製造後如何有效評估大規模機台群的製程速率亦是一大課題;若是引用巨量資料與機器學習的技術,便可幫助工程師做出初步的預判,藉此降低工程師的設計時間,並節省製程及人力成本。

]]>
Thu, 6 Dec 2018 11:08:40 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2318/
Statistical and Computational Approaches for the Identification of Novel Viruses and Virus-host Interactions http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2308/

Abstract


    Viruses play important roles in controlling bacterial population size, altering host metabolism, and have broader impacts on the functions of microbial communities, such as human gut, soil, and ocean microbiomes. However, the investigations of viruses and their functions were vastly underdeveloped. Metagenomic studies provide enormous resources for the identifications of novel viruses and their hosts. We recently developed a k-mer based method, VirFinder, for the identification of novel virus contigs in metagenomic samples [1]. Applications to a liver cirrhosis metagenomic data suggest that viruses play important roles in the development of the disease. We also developed an alignment-free statistic, VirHost-Matcher, for the identification of bacterial hosts of viruses [2] and machine learning based approaches to identify new viruses infecting particular hosts with a relative large number of infecting viruses [3]. Recently we also developed an integrative approach for predicting virus-host interactions [4].

 

1.    J Ren, NA Ahlgren, et al. (2017) Microbiome 5(1):69

2.    NA Ahlgren,  J Ren, et al. (2017) Nucleic Acids Research 45(1):39-53

3.    MG Zhang, et al. (2017) BMC Bioinformatics (APBC2017) 18 (3), 60

4.    WL Wang, J Ren et al. (2018) Under Review.

]]>
Fri, 19 Oct 2018 09:27:19 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2308/
Notes on power comparison between the sequential parallel comparison and other commonly-used designs http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2315/

Abstract

    Because one excludes in the sequential parallel comparison design (SPCD) both placebo responders and patients assigned to the experimental treatment in period 1 from comparison in period 2, test procedures for the SPCD may lose efficiency.  We evaluate the loss of efficiency by comparing power of the SPCD with those for the parallel groups design with repeated measurements (PGDRM), the simple crossover design (SCD), and the parallel group design (PGD). Using Monte Carlo simulations, we demonstrate that one can increase the efficiency of the SPCD by use of the PGDRM and SCD. However, the SPCD may be of use if we wish to assess the treatment effect under the simple carry-over model.

]]>
Mon, 19 Nov 2018 19:43:46 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2315/
A Simple and Efficient Estimation of the Average Treatment Effect in the Presence of Unmeasured Confounders http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2316/

Abstract

    Wang and Tchetgen Tchetgen (2017) studied identification and estimation of the average treatment effect when some confounders are unmeasured. Under their identification condition, they showed that the semiparametric efficient influence function depends on five unknown functionals. They proposed to parameterize all functionals and estimate the average treatment effect from the efficient influence function by replacing the unknown functionals with estimated functionals. They established that their estimator is consistent when certain functionals are correctly specified and attains the semiparametric efficiency bound when all functionals are correctly specified. In applications, it is likely that those functionals could all be misspecified. Consequently their estimator could be inconsistent or consistent but not efficient. This paper presents an alternative estimator that does not require parameterization of any of the functionals. We establish that the proposed estimator is always consistent and always attains the semiparametric efficiency bound. A simple and intuitive estimator of the asymptotic variance is presented, and a small scale simulation study reveals that the proposed estimation outperforms the existing alternatives in finite samples.

]]>
Tue, 27 Nov 2018 10:07:56 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2316/
The de-biased group Lasso estimation for varying coefficient models http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2314/

Abstract

    There has been a lot of attention on the de-biased or de-sparsified Lasso since it was proposed in 2014. The Lasso is very useful in variable selection and obtaining initial estimators for other methods in high-dimensional settings. However, it is well-known that the Lasso produces biased estimators. Therefore several authors simultaneously proposed the de-biased Lasso to fix this drawback and carry out statistical inferences based on the de-biased Lasso estimators. The de-biased Lasso procedures need desirable estimators of high-dimensional precision matrices for bias correction. Thus the research is almost limited to linear regression models with some restrictive assumptions, generalized linear models with stringent assumptions and the like. To our knowledge, there are a few papers on linear regression models with group structure, but no result on structured nonparametric regression models such as varying coefficient models. In this paper, we apply the de-biased group Lasso to varying coefficient models and closely examine the theoretical properties and the effects of approximation errors involved in nonparametric regression. Some simulation results are also presented.

]]>
Mon, 19 Nov 2018 19:38:07 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2314/