Table 1 Simulated data from two independent samples (Group 1 and Group 2) were used in the modelling Cases 1A-1C to determine if their population distributions differ.
From: Rethinking ratio-based normalization towards model-based approaches in heart weight analysis
Simulation case | Group | Group size | Parameter 1 | Parameter 2 | Error term ε |
|---|---|---|---|---|---|
Case 1A | Group 1 | n=10 | N(35,3) | 30 + (1.5 * Parameter 1) + ε | N(0,3) |
Group 2 | n=10 | N(30,3) | 30 + (1.5 * Parameter 1) + ε | N(0,3) | |
Case 1B | Group 1 | n=10 | N(35,3) | 1.5 * Parameter 1 + ε | N(0,3) |
Group 2 | n=10 | N(30,3) | 1.5 * Parameter 1 + ε | N(0,3) | |
Case 1C | Group 1 | n=10 | N(35,3) | N(80,5) | |
Group 2 | n=10 | N(30,3) | N(80,5) |
Case | How P2 is determined |
|---|---|
1A | Related to P1, but not in a fixed proportion (sophisticated function) |
1B | A simple multiple of P1 (proportional) |
1C | Random, with no connection to P1 (independent) |
