Daily update | 13 July, 2022
StackExchange
What is the difference between a logistic regression and log binomial regression?

Source: stats
Views: 336
Score: 7
Tags: regression logistic
Loss function and model parameters estimation and link it to optimization

Source: stats
Views: 26
Score: 2
Tags: regression optimization loss-functions
Why are $R^2$ values not available in mixed effects meta regressions

Source: stats
Views: 14
Score: 2
Tags: r meta-analysis meta-regression metafor meta-learning
What could go wrong if I do gradient descent by class?

Source: stats
Views: 17
Score: 2
Tags: machine-learning neural-networks optimization gradient-descent stochastic-gradient-descent
Minimum sample size calculations for two-sample two-tailed proportion z test

Source: stats
Views: 87
Score: 2
Tags: p-value sample-size z-test survey-sampling