Skip to main content
4 votes

Sample Size for Logistic Regression

The minimum sample size just to do the analysis at all, depends on the number of parameters estimated, so for categorical variables you should count the number of dummy variables. The actual required ...
Frans Rodenburg's user avatar
4 votes

Why should you want to use a smaller control group?

With multiple treatment arms that are being compared to a single control arm, the logic is a bit different than usually. In this situation, not a 1:1:...:1 allocation, but the square root allocation $...
LuckyPal's user avatar
  • 1,900
2 votes

Considering 96 observation for estimating the intercept (rule of thumb)

As I recall, that recommendation for 96 is to get a precise estimate of the intercept (log-odds at reference conditions) in a logistic regression model. If all you care about is how changes in ...
EdM's user avatar
  • 95.6k
2 votes

How to do a power analysis on density (count/area) data?

The first step to start to do what you want is to fit an appropriate model to all the count data that you have for that species, over all years. Working with "densities" ignores the ...
EdM's user avatar
  • 95.6k
1 vote

Sample size calculation clinical trial

A percentage reduction can mean many things - it can mean effects for analyses of continuous outcomes on the logarithmic scale (like 70% off a sticker price). But because no standard deviation is ...
AdamO's user avatar
  • 63.7k
1 vote
Accepted

How to do a power analysis on density (count/area) data?

Overview Simulation is a good way to estimate power in situations that aren't covered by simple formulas. In your case, use your current data to estimate the characteristics of the data generating ...
EdM's user avatar
  • 95.6k
1 vote

How to know what sample size (number of cells) is sufficient to determine what percentage of cells are above a threshold?

To validate the threshold for cell proliferation you have a tough road ahead. You need to show that if a cell count is below the threshold it doesn’t matter biologically how far below the threshold ...
Frank Harrell's user avatar
1 vote
Accepted

discrete time model and Logistic regression

In reverse order: Question 2. The power of a survival model is a function of the number of events; that of logistic regression is a function of the number of cases in the minority class. In both ...
EdM's user avatar
  • 95.6k

Only top scored, non community-wiki answers of a minimum length are eligible