Table 7. Predicted completeness and Queiroz et al (2017) estimates of completeness by state of residence (%), Brazil, 2000–2010, both sexes, ages 5+.
| State | Model 1 | GGB | Model 1 minus GGB | Hybrid | Model 1 minus Hybrid |
|---|---|---|---|---|---|
| Rondônia | 95 | 96 | -1 | 92 | 4 |
| Acre | 93 | 94 | -1 | 87 | 6 |
| Amazonas | 93 | 99 | -6 | 91 | 2 |
| Roraima | 94 | 100 | -6 | 85 | 9 |
| Pará | 91 | 80 | 11 | 77 | 14 |
| Amapá | 92 | 95 | -3 | 84 | 9 |
| Tocantins | 89 | 96 | -7 | 85 | 4 |
| Maranhão | 75 | 90 | -15 | 71 | 4 |
| Piauí | 89 | 98 | -9 | 87 | 2 |
| Ceará | 88 | 99 | -11 | 87 | 0 |
| Rio Grande do Norte | 92 | 99 | -7 | 89 | 3 |
| Paraíba | 95 | 98 | -3 | 90 | 5 |
| Pernambuco | 97 | 100 | -4 | 95 | 2 |
| Alagoas | 95 | 99 | -4 | 92 | 3 |
| Sergipe | 95 | 101 | -6 | 93 | 2 |
| Bahia | 88 | 98 | -10 | 89 | -1 |
| Minas Gerais | 95 | 100 | -5 | 93 | 2 |
| Espírito Santo | 96 | 107 | -11 | 99 | -3 |
| Rio de Janeiro | 99 | 100 | -2 | 96 | 3 |
| São Paulo | 98 | 101 | -3 | 100 | -2 |
| Paraná | 98 | 104 | -7 | 99 | -2 |
| Santa Catarina | 95 | 100 | -5 | 94 | 2 |
| Rio Grande do Sul | 98 | 103 | -4 | 99 | -1 |
| Mato Grosso do Sul | 97 | 107 | -10 | 97 | 0 |
| Mato Grosso | 96 | 100 | -4 | 93 | 3 |
| Goiás | 95 | 99 | -4 | 91 | 4 |
| Distrito Federal | 97 | 98 | -2 | 101 | -4 |
| Mean absolute difference | - | - | -5 | - | 3 |
| Root mean squared difference | - | - | 6 | - | 3 |
GGB: Generalised Growth Balance method. Completeness for 2000–2010 for Model 1 and Model 2 was estimated by making annual estimates of completeness from 2000 to 2010, and weighting by annual completeness by the annual number of registered deaths. GGB and Hybrid estimates of completeness for both sexes were made by weighting sex-specific estimates of completeness in Queiroz et al (Tables 1 and 2) by sex-specific registered deaths [28].