turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Prediction Following a Log Transformed Dependent V...

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-14-2017 04:51 PM

I have the following QLIM tobit model:

Log_Y = intercept + BiX + error

I want to compute the predicted values of Y (y-hat), and I am not sure how to do that. One way is as discussed in this article, but no clear code. www2.sas.com/proceedings/forum2008/370-2008.pdf

Did anyone able to conduct this, or any other method to get the predcited values from the model's prediction of the log?

I know how to do a simple exponentiation of the log, but that way gives me an erroneous prediction (super high compared to the actual).

Any other method can also work for me as long as it has been tested before.

My code is as follows:

```
proc qlim data=WORK.INPUT ;
class Group Location ;
model log_y = Group Location X1 X2 X3;
endogenous log_y ~ censored(lb=0);
output out=OUT1 RESIDUAL PREDICTED;
run;
```

The model gives me the p_log_y, and my question is how to transform this into a "correct" value. Am I doing something wrong?

Thanks.