<P 0pt? 0cm>下面是我之前做的没有加虚拟变量时的程序，但是题目要求加入虚拟变量才能估计出周一，周二<FONT face="Times New Roman">...</FONT>，周五的收益率的参数。这下我可就做不出来了，哪位高手会做的，请帮帮忙啊，多谢拉<FONT face="Times New Roman"> !!!</FONT></P>
<P 0pt? 0cm><o:p><FONT face="Times New Roman"></FONT></o:p></P>
<P 0pt? 0cm><FONT face="Times New Roman">data a;</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">input date:date9. r;</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">cards;</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">5Jan93      0.053433404</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">6Jan93      -0.013222868</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">7Jan93       0.025088562</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">8Jan93       0.023317177</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">11Jan93     0.094290667</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">12Jan93     0.010016982</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">13Jan93     0.036587203</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">14Jan93     0.059131738</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">15Jan93     -0.02108729</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">18Jan93     -0.007965763</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">19Jan93     -0.062047596</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">20Jan93     0.050601909</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">21Jan93     0.029233701</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">    ...            ...</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">  proc autoreg data=a;</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">model r=date/nlag=2 garch=(q=1,p=1) maxit=50;</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman">run;</FONT></P>
<P 0pt? 0cm><FONT face="Times New Roman"></FONT></P>

[此贴子已经被作者于2007-3-12 1:11:28编辑过]

<P>不知道哪位熟悉GARCH的高手，能帮我看看怎么做啊？</P>
<P>多谢拉！</P>

the following text is tranferred from the sas mannual, you may find the way if you check the detail.<div><br/><div>Using the HETERO Statement with GARCH ModelsThe HETERO statement can be combined
with the GARCH=
option on the MODEL statement to include input variables in the GARCH conditional
variance model. For example, the GARCH(1,1) variance model with two dummy input variables D1
and D2 is


The following statements estimate this GARCH model:<div><br/></div><div>proc autoreg data=one;
      model y = x z / garch=(p=1,q=1);
      hetero d1 d2;
   run;</div><div><br/></div><div> The parameters for the variables D1 and D2 can be constrained using the COEF=
option. For example, the constraints  are imposed by the following statements:</div><div> proc autoreg data=one;
      model y = x z / garch=(p=1,q=1);
      hetero d1 d2 / coef=unit;
   run;</div></div></div>

good to learn. thanks.<br/>

thanks for sharing