| 1. | :It sounds like you want Sequential probability ratio test ( SPRT ).
|
| 2. | Extensions to ratio test, however, sometimes allows one to deal with this case.
|
| 3. | Does the ratio test only work for series and not sequences?
|
| 4. | Comparison functions are necessarily entire, which follows from the ratio test.
|
| 5. | That is quite different from doing any sort of ratio test.
|
| 6. | The models being compared need not be F-test or a likelihood ratio test.
|
| 7. | If the ratio test is inconclusive, and the series may converge or diverge.
|
| 8. | A general approach in a frequentist analysis can be based on maximum likelihood-ratio tests.
|
| 9. | With sequences, the ratio test only determines convergence to zero.
|
| 10. | One such statistic is the chi-square likelihood-ratio test, which assesses the difference between models.
|