Dataset Information

Review and evaluation of penalised regression methods for risk prediction in low-dimensional data with few events.

ABSTRACT: Risk prediction models are used to predict a clinical outcome for patients using a set of predictors. We focus on predicting low-dimensional binary outcomes typically arising in epidemiology, health services and public health research where logistic regression is commonly used. When the number of events is small compared with the number of regression coefficients, model overfitting can be a serious problem. An overfitted model tends to demonstrate poor predictive accuracy when applied to new data. We review frequentist and Bayesian shrinkage methods that may alleviate overfitting by shrinking the regression coefficients towards zero (some methods can also provide more parsimonious models by omitting some predictors). We evaluated their predictive performance in comparison with maximum likelihood estimation using real and simulated data. The simulation study showed that maximum likelihood estimation tends to produce overfitted models with poor predictive performance in scenarios with few events, and penalised methods can offer improvement. Ridge regression performed well, except in scenarios with many noise predictors. Lasso performed better than ridge in scenarios with many noise predictors and worse in the presence of correlated predictors. Elastic net, a hybrid of the two, performed well in all scenarios. Adaptive lasso and smoothly clipped absolute deviation performed best in scenarios with many noise predictors; in other scenarios, their performance was inferior to that of ridge and lasso. Bayesian approaches performed well when the hyperparameters for the priors were chosen carefully. Their use may aid variable selection, and they can be easily extended to clustered-data settings and to incorporate external information.

SUBMITTER: Pavlou M

PROVIDER: S-EPMC4982098 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundPenalised regression methods are a useful atheoretical approach for both developing predictive models and selecting key indicators within an often substantially larger pool of available indicators. In comparison to traditional methods, penalised regression models improve prediction in new data by shrinking the size of coefficients and retaining those with coefficients greater than zero. However, the performance and selection of indicators depends on the specific algorithm implemented. The purpose of this study was to examine the predictive performance and feature (i.e., indicator) selection capability of common penalised logistic regression methods (LASSO, adaptive LASSO, and elastic-net), compared with traditional logistic regression and forward selection methods.DesignData were drawn from the Australian Temperament Project, a multigenerational longitudinal study established in 1983. The analytic sample consisted of 1,292 (707 women) participants. A total of 102 adolescent psychosocial and contextual indicators were available to predict young adult daily smoking.FindingsPenalised logistic regression methods showed small improvements in predictive performance over logistic regression and forward selection. However, no single penalised logistic regression model outperformed the others. Elastic-net models selected more indicators than either LASSO or adaptive LASSO. Additionally, more regularised models included fewer indicators, yet had comparable predictive performance. Forward selection methods dismissed many indicators identified as important in the penalised logistic regression models.ConclusionsAlthough overall predictive accuracy was only marginally better with penalised logistic regression methods, benefits were most clear in their capacity to select a manageable subset of indicators. Preference to competing penalised logistic regression methods may therefore be guided by feature selection capability, and thus interpretative considerations, rather than predictive performance alone.

Project description:DNA microarray studies identified gene expression signatures predictive of metastatic relapse in early breast cancer. Standard feature selection procedures applied to reduce the set of predictive genes did not take into account the correlation between genes. In this paper, we studied the performances of three high-dimensional regression methods - CoxBoost, LASSO (Least Absolute Shrinkage and Selection Operator), and Elastic net - to identify prognostic signatures in patients with early breast cancer.We analyzed three public retrospective datasets, including a total of 384 patients with axillary lymph node-negative breast cancer. The Amsterdam van't Veer's training set of 78 patients was used to determine the optimal gene sets and classifiers using sensitivity thresholds resulting in mis-classification of no more than 10% of the poor-prognosis group. To ensure the comparability between different methods, an automatic selection procedure was used to determine the number of genes included in each model. The van de Vijver's and Desmedt's datasets were used as validation sets to evaluate separately the prognostic performances of our classifiers. The results were compared to the original Amsterdam 70-gene classifier.The automatic selection procedure reduced the number of predictive genes up to a minimum of six genes. In the two validation sets, the three models (Elastic net, LASSO, and CoxBoost) led to the definition of genomic classifiers predicting the 5-year metastatic status with similar performances, with respective 59, 56, and 54% accuracy, 83, 75, and 83% sensitivity, and 53, 52, and 48% specificity in the Desmedt's dataset. In comparison, the Amsterdam 70-gene signature showed 45% accuracy, 97% sensitivity, and 34% specificity. The gene overlap and the classification concordance between the three classifiers were high. All the classifiers added significant prognostic information to that provided by the traditional prognostic factors and showed a very high overlap with respect to gene ontologies (GOs) associated with genes overexpressed in the predicted poor-prognosis vs. good-prognosis classes and centred on cell proliferation. Interestingly, all classifiers reported high sensitivity to predict the 4-year status of metastatic disease.High-dimensional regression methods are attractive in prognostic studies because finding a small subset of genes may facilitate the transfer to the clinic, and also because they strengthen the robustness of the model by limiting the selection of false-positive predictive genes. With only six genes, the CoxBoost classifier predicted the 4-year status of metastatic disease with 93% sensitivity. Selecting a few genes related to ontologies other than cell proliferation might further improve the overall sensitivity performance.

Dataset Information

Review and evaluation of penalised regression methods for risk prediction in low-dimensional data with few events.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets