全部版块 我的主页
论坛 提问 悬赏 求职 新闻 读书 功能一区 经管百科 爱问频道
1814 5
2014-08-13
能不能建立一个 一元二次 方程的线性方程 去验证一个结论
比如说结论是一个变量时关于另一个变量的散点图是抛物线形状
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

全部回复
2014-8-13 10:38:29
顶一顶啊啊
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

2014-8-13 10:49:31
这意思是可以么
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

2014-8-13 10:54:38
Hello buddy,
normally there is no one prohibit you from doing that.

But as a practitioner, I seldom to do so.
More often,  we use lineal equation to handle a business issue.

If your conclusion is beyond my description,
just do it,
as long as you can illustrate it clearly,
you are good.

As far as I know,
there is no restriction on your methodology.
Good luck again.
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

2014-8-13 20:39:22
cpamodeler 发表于 2014-8-13 10:54
Hello buddy,
normally there is no one prohibit you from doing that.
我能不能再问您个问题,VAR模型做完johansan协整检验,怎么看协整方程
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

2014-8-13 20:51:41
Hi buddy,
I must admit that you mentioned something I don't know.

But after searching my relevant experience,
I made a bold guess.

I remember that my professor told me:
I should evaluate the correlation between those independent variables.
If there was a strong correlation between them,
you'd better at least get rid of one.

Or you can delete another one in your next model.

After comparing the p-value, keep the best one.

I don't know whether this is what you want,
just do whatever I could to save your time.
Thank you.
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

相关推荐
栏目导航
热门文章
推荐文章

说点什么

分享

扫码加好友,拉您进群
各岗位、行业、专业交流群