https://www.facebook.com/cienciasinseso/

Sin categoría / Statistics

The Palace of Probabilities

With small sample size, the variance of the Pearson’s coefficient increases, decreasing the precision of its estimates. The use of Fisher's z helps to stabilize the variance and obtain more precise estimates in meta-analyses whose outcome measure is a correlation, and when the sample of primary studies is small.

02/27/2024

Epidemiology / Statistics

The art of resignation

The procedure for choosing the cut-off point for a diagnostic test is reviewed. To decide this threshold, which is influenced by the characteristics of the model and the clinical scenario in which it will be applied, we will take into account the sensitivity and precision of the test for each possible cut-off point. The precision enrichment ratio will be useful in cases with a large imbalance between the two diagnostic categories.

01/30/2024

Sin categoría / Statistics

An epic dance

Multiple regression regularization (shrinkage) techniques can be very useful to address collinearity or overfitting problems. In addition, they can be used to select the independent variables and reduce multidimensionality, achieving more robust and easy-to-interpret models. Ridge, lasso and elastic network regression techniques are described.

12/05/2023

Epidemiology / Machine learning

An intruder from another world

The F1-score, also called F-score or F-measure, is an estimator of the classification capacity of a test that is frequently used in data science and artificial intelligence algorithms and that can be useful for evaluation of diagnostic tests. It is the harmonic mean of sensitivity and positive predictive value, so it weights the value of both in a single estimator.

11/06/2023

Sin categoría / Statistics

The paradox of the air

The parameters that report on the quality of a multiple logistic regression model and that are usually provided by the statistical programs with which they are carried out are reviewed. Emphasis is placed on the goodness of fit, the predictive capacity and the statistical significance of the model.

09/25/2023