面板数据分析-经管之家官网！

免费学术公开课,扫码加入

用STATA做PANEL时,用的是固定效果模型,因变量是一个序次变量(0 很困难 1 比较困难 2 一般 3 比较富裕 4 很富裕),用xtlogit时出现

note: multiple positive outcomes within groups encountered.
note: 2258 groups (4070 obs) dropped due to all positive or
all negative outcomes.

Iteration 0: log likelihood = -20.415268 (not concave)
Iteration 1: log likelihood = -18.280986 (not concave)
Iteration 2: log likelihood = -17.908176 (not concave)
Iteration 3: log likelihood = -17.745828 (not concave)
Iteration 4: log likelihood = -17.662535 (not concave)
Iteration 5: log likelihood = -17.584351 (not concave)
Iteration 6: log likelihood = -17.552826 (not concave)
Iteration 7: log likelihood = -17.530743 (not concave)
Iteration 8: log likelihood = -17.511968 (not concave)
Iteration 9: log likelihood = -17.493531 (not concave)
Iteration 10: log likelihood = -17.484152 (not concave)
Iteration 11: log likelihood = -17.41418 (not concave)
Iteration 12: log likelihood = -17.40963 (not concave)
Iteration 13: log likelihood = -17.406726 (not concave)
Iteration 14: log likelihood = -17.40411 (not concave)
Iteration 15: log likelihood = -17.399807 (not concave

而且一直迭代没有停

请教大家这是什么意思啊

「经管之家」APP：经管人学习、答疑、交友，就上经管之家！
免流量费下载资料----在经管之家app可以下载论坛上的所有资源，并且不额外收取下载高峰期的论坛币。
涵盖所有经管领域的优秀内容----覆盖经济、管理、金融投资、计量统计、数据分析、国贸、财会等专业的学习宝库，各类资料应有尽有。
来自五湖四海的经管达人----已经有上千万的经管人来到这里，你可以找到任何学科方向、有共同话题的朋友。
经管之家（原人大经济论坛），跨越高校的围墙，带你走进经管知识的新世界。
扫描下方二维码下载并注册APP