<p>各位大侠好：</p><p>本人新手上路，因为时间的关系，所以我没有办法系统地查看计量经济学与stata教程，在统计数据做logit模型时遇到几个问题想请教各位高手，如若能解答感激不尽，具体问题如下：</p><p>1.有些文章中，在用stata做logit模型时，关于“年龄（周岁）”这个解释变量，统计结果不仅列出了“年龄”本身的回归系数及显著程度，还列出了“年龄的平方项”的回归系数及显著程度，请问这个“年龄的平方项”有什么作用呢？</p><p>2.关于定序变量，如“收入水平INCOME”，我的数据中分为：1.1000及以下；2.1001-2500；3.2501-5000；4.5001及以上。我想将2、3、4及个等级的INCOME分别与1进行比较，也就是说参照组是1000及以下。比如在将3与1比较时，所有选3的数据均录入为1，那么我应该只将选1的数据录入为0还是应该将所有非3的数据均录入为0？</p><p>3.用STATA做logit模型时，应该怎么求边际效应?比如：因变量Y（1=购买，0=不购买），解释变量sex（1=女，0=男），education(<span style="; FONT-FAMILY: 宋体; mso-ascii-font-family: "Times New Roman"; mso-bidi-font-family: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;">初中及以下</span><span lang="EN-US" style="; FONT-FAMILY: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-font-family: 宋体;">=1</span><span style="; FONT-FAMILY: 宋体; mso-ascii-font-family: "Times New Roman"; mso-bidi-font-family: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;">；普通高中和中专</span><span lang="EN-US" style="; FONT-FAMILY: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-font-family: 宋体;">=2</span><span style="; FONT-FAMILY: 宋体; mso-ascii-font-family: "Times New Roman"; mso-bidi-font-family: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;">；大专</span><span lang="EN-US" style="; FONT-FAMILY: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-font-family: 宋体;">=3</span><span style="; FONT-FAMILY: 宋体; mso-ascii-font-family: "Times New Roman"; mso-bidi-font-family: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;">；本科</span><span lang="EN-US" style="; FONT-FAMILY: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-font-family: 宋体;">=4</span><span style="; FONT-FAMILY: 宋体; mso-ascii-font-family: "Times New Roman"; mso-bidi-font-family: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA;">；研究生及以上</span><span lang="EN-US" style="; FONT-FAMILY: "Times New Roman"; mso-font-kerning: 1.0pt; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-font-family: 宋体;">=5</span>),想要知道购买概率女性会比男性高多少或是低多少？在平均值附近的教育水平每增加一级，购买的概率会变化多少？这样的问题该如何求？</p>

[此贴子已经被作者于2008-5-9 10:26:14编辑过]