Simulation: case parameters and accuracy for competing methods.
| Methods: mean accuracy (SE) | ||||||
|---|---|---|---|---|---|---|
| Case | Simulation | Parameters | PMM | DT-HMM | CT-HMM | PH-HMM |
| 1.1 |  |  | 0.985(0.003) | 0.990(0.003) | 0.983(0.003) | 0.984(0.004) |
| 1.2 |  |  | 0.985(0.004) | 0.999(0.001) | 0.998(0.001) | 0.998(0.001) |
| 1.3 |  |  | 0.985(0.003) | 0.997(0.001) | 0.997(0.001) | 0.997(0.001) |
| 2.1 |  |  | 0.985(0.003) | 0.994(0.002) | 0.979(0.005) | 0.991(0.003) |
| 2.2 |  |  | 0.984(0.026) | 0.998(0.001) | 0.997(0.001) | 0.998(0.001) |
| 2.3 |  |  | 0.985(0.003) | 0.997(0.002) | 0.997(0.002) | 0.997(0.002) |
| 3.1 |  |  | 0.898(0.008) | 0.926(0.008) | 0.896(0.010) | 0.873(0.015) |
| 3.2 |  |  | 0.894(0.034) | 0.988(0.004) | 0.986(0.004) | 0.982(0.007) |
| 3.3 |  |  | 0.896(0.016) | 0.977(0.005) | 0.977(0.005) | 0.977(0.005) |
| 4.1 |  |  | 0.897(0.019) | 0.964(0.005) | 0.801(0.022) | 0.904(0.013) |
| 4.2 |  |  | 0.897(0.009) | 0.986(0.004) | 0.984(0.005) | 0.978(0.007) |
| 4.3 |  |  | 0.897(0.013) | 0.977(0.005) | 0.975(0.005) | 0.974(0.005) |
| Methods: mean accuracy (SE) | ||||||
|---|---|---|---|---|---|---|
| Case | Simulation | Parameters | PMM | DT-HMM | CT-HMM | PH-HMM |
| 1.1 | | | 0.985(0.003) | 0.990(0.003) | 0.983(0.003) | 0.984(0.004) |
| 1.2 | | | 0.985(0.004) | 0.999(0.001) | 0.998(0.001) | 0.998(0.001) |
| 1.3 | | | 0.985(0.003) | 0.997(0.001) | 0.997(0.001) | 0.997(0.001) |
| 2.1 | | | 0.985(0.003) | 0.994(0.002) | 0.979(0.005) | 0.991(0.003) |
| 2.2 | | | 0.984(0.026) | 0.998(0.001) | 0.997(0.001) | 0.998(0.001) |
| 2.3 | | | 0.985(0.003) | 0.997(0.002) | 0.997(0.002) | 0.997(0.002) |
| 3.1 | | | 0.898(0.008) | 0.926(0.008) | 0.896(0.010) | 0.873(0.015) |
| 3.2 | | | 0.894(0.034) | 0.988(0.004) | 0.986(0.004) | 0.982(0.007) |
| 3.3 | | | 0.896(0.016) | 0.977(0.005) | 0.977(0.005) | 0.977(0.005) |
| 4.1 | | | 0.897(0.019) | 0.964(0.005) | 0.801(0.022) | 0.904(0.013) |
| 4.2 | | | 0.897(0.009) | 0.986(0.004) | 0.984(0.005) | 0.978(0.007) |
| 4.3 | | | 0.897(0.013) | 0.977(0.005) | 0.975(0.005) | 0.974(0.005) |
Note: Mean accuracy and empirical standard errors based on 500 replicates. Each replicate had 
individuals with 
state transitions. The survival models were simulate with Algorithm 1 and logistic models were simulate with Algorithm 2 from the Web Appendix. Cases 1.1–1.3 looked at low incidence rate and large distributional difference; Cases 2.1–2.3 looked at high incidence rate and large distributional difference; Cases 3.1–3.3 looked at low incidence rate and small distributional difference; and Cases 4.1–4.3 looked at high incidence rate and small distributional difference.
Simulation: case parameters and accuracy for competing methods.
| Methods: mean accuracy (SE) | ||||||
|---|---|---|---|---|---|---|
| Case | Simulation | Parameters | PMM | DT-HMM | CT-HMM | PH-HMM |
| 1.1 | | | 0.985(0.003) | 0.990(0.003) | 0.983(0.003) | 0.984(0.004) |
| 1.2 | | | 0.985(0.004) | 0.999(0.001) | 0.998(0.001) | 0.998(0.001) |
| 1.3 | | | 0.985(0.003) | 0.997(0.001) | 0.997(0.001) | 0.997(0.001) |
| 2.1 | | | 0.985(0.003) | 0.994(0.002) | 0.979(0.005) | 0.991(0.003) |
| 2.2 | | | 0.984(0.026) | 0.998(0.001) | 0.997(0.001) | 0.998(0.001) |
| 2.3 | | | 0.985(0.003) | 0.997(0.002) | 0.997(0.002) | 0.997(0.002) |
| 3.1 | | | 0.898(0.008) | 0.926(0.008) | 0.896(0.010) | 0.873(0.015) |
| 3.2 | | | 0.894(0.034) | 0.988(0.004) | 0.986(0.004) | 0.982(0.007) |
| 3.3 | | | 0.896(0.016) | 0.977(0.005) | 0.977(0.005) | 0.977(0.005) |
| 4.1 | | | 0.897(0.019) | 0.964(0.005) | 0.801(0.022) | 0.904(0.013) |
| 4.2 | | | 0.897(0.009) | 0.986(0.004) | 0.984(0.005) | 0.978(0.007) |
| 4.3 | | | 0.897(0.013) | 0.977(0.005) | 0.975(0.005) | 0.974(0.005) |
| Methods: mean accuracy (SE) | ||||||
|---|---|---|---|---|---|---|
| Case | Simulation | Parameters | PMM | DT-HMM | CT-HMM | PH-HMM |
| 1.1 | | | 0.985(0.003) | 0.990(0.003) | 0.983(0.003) | 0.984(0.004) |
| 1.2 | | | 0.985(0.004) | 0.999(0.001) | 0.998(0.001) | 0.998(0.001) |
| 1.3 | | | 0.985(0.003) | 0.997(0.001) | 0.997(0.001) | 0.997(0.001) |
| 2.1 | | | 0.985(0.003) | 0.994(0.002) | 0.979(0.005) | 0.991(0.003) |
| 2.2 | | | 0.984(0.026) | 0.998(0.001) | 0.997(0.001) | 0.998(0.001) |
| 2.3 | | | 0.985(0.003) | 0.997(0.002) | 0.997(0.002) | 0.997(0.002) |
| 3.1 | | | 0.898(0.008) | 0.926(0.008) | 0.896(0.010) | 0.873(0.015) |
| 3.2 | | | 0.894(0.034) | 0.988(0.004) | 0.986(0.004) | 0.982(0.007) |
| 3.3 | | | 0.896(0.016) | 0.977(0.005) | 0.977(0.005) | 0.977(0.005) |
| 4.1 | | | 0.897(0.019) | 0.964(0.005) | 0.801(0.022) | 0.904(0.013) |
| 4.2 | | | 0.897(0.009) | 0.986(0.004) | 0.984(0.005) | 0.978(0.007) |
| 4.3 | | | 0.897(0.013) | 0.977(0.005) | 0.975(0.005) | 0.974(0.005) |
Note: Mean accuracy and empirical standard errors based on 500 replicates. Each replicate had individuals with state transitions. The survival models were simulate with Algorithm 1 and logistic models were simulate with Algorithm 2 from the Web Appendix. Cases 1.1–1.3 looked at low incidence rate and large distributional difference; Cases 2.1–2.3 looked at high incidence rate and large distributional difference; Cases 3.1–3.3 looked at low incidence rate and small distributional difference; and Cases 4.1–4.3 looked at high incidence rate and small distributional difference.
