发帖
雷达卡
1) The underlying relationship between Y and X is Yi=βXi+εi, where the density function ofεi is f(εi)= exp(-εi) for εi non-negative and zero otherwise. The values of X are observed, but Y is an unobserved latent variable. The only thing you know is the value of an indicator variable Z that is 1 when Y is positive and 0 when it is not positive. Using the data below, find the maximum likelihood estimate for β and test the hypothesis that β=5 using a likelihood ratio test.
| X | Z |
| 2.330202 | 1 |
| 0.412245 | 1 |
| -0.95171 | 0 |
| 0.652971 | 1 |
| 1.694773 | 1 |
| 0.11577 | 1 |
| 2.342919 | 1 |
| 2.111553 | 1 |
| -1.45119 | 0 |
| -0.41914 | 0 |
| -1.876 | 0 |
| 1.911599 | 1 |
| -0.4387 | 0 |
| 1.452094 | 1 |
| 1.928657 | 1 |
| -2.48699 | 0 |
| 1.704174 | 1 |
| 0.231499 | 1 |
| -2.403 | 0 |
| 2.293572 | 1 |
| 1.140321 | 1 |
| -0.69274 | 0 |
| -1.09291 | 0 |
| 2.016314 | 1 |
| -0.75442 | 0 |
| -1.84381 | 0 |
| -1.6475 | 0 |
| -2.37047 | 0 |
| -2.35681 | 0 |
| -0.14848 | 0 |
推倒了半天
觉得应该是找个β maximizing L=∏(1-EXP(βXi)) for all Xi<0
请教怎样用matlab做，我不会，我只会eviews
京公网安备 11010802022788号
