Hello,
Thank you in advance.
If I want to report the adjusted (predicted) median price (see example below), adjusted for weight and mpg, which summary statistic should I use?
Also, I know that I can easily do this using the 'adjust' command as well, but my question is: If I wanted to report the adjusted median, do I want to report the median (50%ile) or the mean "of the predicted median"?
See example:
---begin---
sysuse auto, clear
qreg price weight mpg
predict adjustedprice, xb
sum adjustedprice, d
---end---
. sum adjustedprice, d
Linear prediction
-------------------------------------------------------------
Percentiles Smallest
1% 3364.087 3364.087
5% 3754.391 3673.913
10% 4097.478 3737.043 Obs 74
25% 4592.348 3754.391 Sum of Wgt. 74
50% 5311.543 Mean 5193.156
Largest Std. Dev. 780.0377
75% 5792.217 6303
90% 6129.522 6418.652 Variance 608458.9
95% 6303 6772.348 Skewness -.2198667
99% 6841.739 6841.739 Kurtosis 2.374105
1. Note that I run my quantile regression model, which essentially generates my predicted median price (adjusting for weight and mpg) (the qreg command).
2. I then generate the predicted values using linear prediction (the predict command).
3. I then summarize the data.
4. Now, if I want to report predicted medians, do I want to use the 50% (median) of the adjusted price, or the mean of the adjusted price (noting that the qreg command provides predicted medians already)?
Thank you
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
