情景:询问受访者喜欢购买什么样的保温杯————————
1.通过正交设计了14种型号(mode)的保温杯,让受访者从14中选择最可能购买的一种
2.有4种属性来勾勒每个型号的保温杯:材质(塑料/不锈钢)、等级(1-3)、是否通过iso认证、价格(45/90/135)
3.收集受访者个人信息:性别、年龄、教育程度、收入水平
问题:
希望考察消费者对哪一种属性更重视,不同类型的消费者有没有不同的偏好(例如教育水平高的会不会对iso认证更敏感一些)
因此使用混合logit模型:既有随型号而别的自变量,又有不变的个人特征变量。
因变量:是否选择这个型号的保温杯(choice)
自变量:caizi dengji iso price 和 age male educ income
数据:宽型排列
命令:
asclogit choice caizi dengji iso price , case(id) alternatives(mode) base(14) casevars(age male educ income ) nolog
反馈:
note: caizi dropped because of collinearity
note: dengji dropped because of collinearity
note: iso dropped because of collinearity
note: model has collinear variables; convergence may not be achieved
14个产品的属性肯定在每个id中都会重复出现一次啊,请问要怎么解决产品属性出现共线性的问题?求大神给予帮助
部分数据如下,全部数据在附件:
| id | choice | mode | caizi | dengji | iso | price | male | age | educ | income |
1 | 0 | 1 | 1 | 3 | 0 | 135 | 0 | 37 | 3 | 2 |
1 | 0 | 2 | 1 | 2 | 1 | 90 | 0 | 37 | 3 | 2 |
1 | 0 | 3 | 0 | 3 | 0 | 90 | 0 | 37 | 3 | 2 |
1 | 0 | 4 | 1 | 1 | 0 | 135 | 0 | 37 | 3 | 2 |
1 | 0 | 5 | 0 | 1 | 1 | 135 | 0 | 37 | 3 | 2 |
1 | 0 | 6 | 0 | 1 | 0 | 45 | 0 | 37 | 3 | 2 |
1 | 0 | 7 | 0 | 2 | 1 | 135 | 0 | 37 | 3 | 2 |
1 | 0 | 8 | 0 | 2 | 0 | 45 | 0 | 37 | 3 | 2 |
1 | 0 | 9 | 1 | 1 | 1 | 90 | 0 | 37 | 3 | 2 |
1 | 0 | 10 | 1 | 1 | 1 | 45 | 0 | 37 | 3 | 2 |
1 | 0 | 11 | 0 | 3 | 1 | 45 | 0 | 37 | 3 | 2 |
1 | 1 | 12 | 1 | 1 | 0 | 45 | 0 | 37 | 3 | 2 |
1 | 0 | 13 | 0 | 1 | 1 | 45 | 0 | 37 | 3 | 2 |
1 | 0 | 14 | 1 | 2 | 0 | 45 | 0 | 37 | 3 | 2 |
2 | 0 | 1 | 1 | 3 | 0 | 135 | 0 | 29 | 3 | 2 |
2 | 0 | 2 | 1 | 2 | 1 | 90 | 0 | 29 | 3 | 2 |
2 | 0 | 3 | 0 | 3 | 0 | 90 | 0 | 29 | 3 | 2 |
2 | 0 | 4 | 1 | 1 | 0 | 135 | 0 | 29 | 3 | 2 |
2 | 0 | 5 | 0 | 1 | 1 | 135 | 0 | 29 | 3 | 2 |
2 | 0 | 6 | 0 | 1 | 0 | 45 | 0 | 29 | 3 | 2 |
2 | 0 | 7 | 0 | 2 | 1 | 135 | 0 | 29 | 3 | 2 |
2 | 0 | 8 | 0 | 2 | 0 | 45 | 0 | 29 | 3 | 2 |
2 | 0 | 9 | 1 | 1 | 1 | 90 | 0 | 29 | 3 | 2 |
2 | 0 | 10 | 1 | 1 | 1 | 45 | 0 | 29 | 3 | 2 |
2 | 0 | 11 | 0 | 3 | 1 | 45 | 0 | 29 | 3 | 2 |
2 | 1 | 12 | 1 | 1 | 0 | 45 | 0 | 29 | 3 | 2 |
2 | 0 | 13 | 0 | 1 | 1 | 45 | 0 | 29 | 3 | 2 |
2 | 0 | 14 | 1 | 2 | 0 | 45 | 0 | 29 | 3 | 2 |