Max-linear regression models with regularization

Cui, Qiurong; Xu, Yuqing; Zhang, Zhengjun; Chan, V. S.

doi:10.1016/j.jeconom.2020.07.017

Cited by 25 publications

(36 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, compared with different classifiers with dozens of genes as predictors, the five-gene-based competing classifier is the most sparse gene-based classifier. Moreover, as discussed in Cui et al (2020) that the classical linear regression is a particular case of the competing factor regression; the classical logistic model is a particular case of the competing factor classifier. In terms of the competing factor classifier itself, if the number of competing factors or the number of gene predictors can be reduced, the five-gene-based competing factor classifier is overfitting the data.…”

Section: Discussionmentioning

confidence: 99%

“…The max-linear competing factor models (Cui and Zhang, 2018), the max-linear regression models (Cui et al, 2020), and the max-linear logistic models (Xu, 2019) have an advantage over existing models in a large class of research problems, e.g., nonlinear predictions and classifications. These models are different from the random forest, support vector machine, group lasso based machine learning methods, and deep learning methods.…”

Section: Statistical Methodologymentioning

confidence: 99%

“…Max-linear models are not only interpretable but also outperform existing methods. In the literature, theoretical statistical and probabilistic foundations related to the competing risk factor models have been established (Cui et al, 2020;Cui and Zhang, 2018;Malinowski et al, 2016;Xu, 2019;Cao and Zhang, 2020;Zhang, 2005Zhang, , 2020. The difference between the max-linear logistic regression and the classical logistic regression is that the original linear combination of predictors is replaced by the maximum of several linear combinations of predictors, called competing factors or competing-risk factors.…”

Section: Statistical Methodologymentioning

confidence: 99%

See 2 more Smart Citations

Five Critical Genes Related to Seven COVID-19 Subtypes: A Data Science Discovery

Zhang¹

2021

Journal of Data Science

Self Cite

129

View full text Add to dashboard Cite

Since the first confirmed case of COVID-19 was identified in December 2019, the total COVID-19 patients are up to 80,675,745, and the number of deaths is 1,764,185 as of December 27, 2020. The problem is that researchers are still learning about it, and new variants of SARS-CoV-2 are not stopping. For medical treatment, essential and informative genes can lead to accurate tests of whether an individual has contracted COVID-19 and help develop highly efficient vaccines, antiviral drugs, and treatments. As a result, identifying critical genes related to COVID-19 has been an urgent task for medical researchers. We conducted a competing risk analysis using the max-linear logistic regression model to analyze 126 blood samples from COVID-19-positive and COVID-19-negative patients. Our research led to a competing COVID-19 risk classifier derived from 19,472 genes and their differential expression values. The final classifier model only involves five critical genes, ABCB6, KIAA1614, MND1, SMG1, RIPK3, which led to 100% sensitivity and 100% specificity of the 126 samples. Given their 100% accuracy in predicting COVID-19 positive or negative status, these five genes can be critical in developing proper, focused, and accurate COVID-19 testing procedures, guiding the second-generation vaccine development, studying antiviral drugs and treatments. It is expected that these five genes can motivate numerous new COVID-19 researches.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Statistical Methodologymentioning

confidence: 99%

Section: Statistical Methodologymentioning

confidence: 99%

See 1 more Smart Citation

Five Critical Genes Related to Seven COVID-19 Subtypes: A Data Science Discovery

Zhang¹

2021

Journal of Data Science

Self Cite

129

View full text Add to dashboard Cite

show abstract

“…The max-linear competing factor models are different from the popular models mentioned earlier. The max-linear competing factor models are interpretable and outperform existing methods (e.g., random forest and graphical group lasso) in estimation accuracy and prediction power under broad data structures 19 . For the theoretical foundation of these new models, we refer the readers to papers 19;18;21;20;22;23 .…”

Section: The Algorithmmentioning

confidence: 92%

“…The most recently developed machine learning methods: max-linear competing factor models 18 , maxlinear regression models 19 , and max-linear logistic models 20;17 , have proven to be a widely applicable class of new models in statistical analysis and max-linear machine learning. The difference between the max-linear competing models and the classical statistical models is that the original linear combination of predictors is replaced by the maximum of several linear combinations of predictors, called competing factors or competingrisk factors.…”

Section: The Algorithmmentioning

confidence: 99%

Functional Effects of Four or Fewer Critical Genes Linked to Lung Cancers and New Subtypes Detected by A New Machine Learning Classifier

Zhang¹

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Finding genes biologically directly or indirectly related to lung cancer has been drawing much attention, and many genes directly related to lung cancer have been reported. However, it has not been confirmed whether those published 'key' genes are truly critical to lung cancer formation, i.e., they may be with very limited useful information. As a result, finding essential genes remains a challenging lung cancer research problem. Using a recently developed competing linear factor analysis method in differentially expressed gene detection, we advance the study of lung cancer critical genes detection to a uniformly informative level. A set of common four genes and their functional effects are detected to be differentially expressed in tumor and non-tumor samples with 100% sensitivity and 100% specificity in one study of lung adenocarcinoma (LUAD) and one study of squamous cell lung cancers (LUSC) (two North American cohorts with 20429 genes, 576 and 552 samples respectively). Two additional analyses also gain accuracy of 97.8% sensitivity and 100% specificity in one study of non-small cell lung carcinomas (NSCLC, a European cohort with 20356 genes and 156 samples), and an accuracy of 100% sensitivity and 95% specificity (1 out of 20 non-tumor samples) in one study of ALK-positive and EGFR/KRAS/ALK-negative lung adenocarcinomas (LUAD, a Japanese cohort with 20356 genes and 224 samples). There are some common genes, but different functional effects, within each set of four genes among two North American cohorts and a European cohort and among North American cohorts and the Japanese cohort. These results show the four-gene-based classifiers are robust with different types of lung cancers and different race cohorts and accurate. The functional effects of four genes disclose significantly other mechanisms (mysteries) between LUAD and LUSC. These sets of four genes and their functional effects are considered to be essential for lung cancer studies and practice. These genes' functional effects naturally classify patients into different groups (more than seven subtypes). Subtype information is useful for personalized therapies. The new findings can motivate new lung cancer research in more focused and targeted directions to save lives, protect people, and reduce enormous economic costs in research and lung cancer treatments.

show abstract

Site-specific nitrogen recommendation: fast, accurate, and feasible Bayesian kriging

Poursina,

Brorsen

2024

Comput Stat

View full text Add to dashboard Cite

Max-linear regression models with regularization

Cited by 25 publications

References 23 publications

Five Critical Genes Related to Seven COVID-19 Subtypes: A Data Science Discovery

Five Critical Genes Related to Seven COVID-19 Subtypes: A Data Science Discovery

Functional Effects of Four or Fewer Critical Genes Linked to Lung Cancers and New Subtypes Detected by A New Machine Learning Classifier

Site-specific nitrogen recommendation: fast, accurate, and feasible Bayesian kriging

Contact Info

Product

Resources

About