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
- /
- value of statistical life

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
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-23-2016 07:06 PM

Does anyone know how to calculate the Value of Statistical life in SAS?

the variables in the data set im working with are defined as:

wage = worker's hourly wage rate

male = 1 if worker is male and =0 otherwise

white = 1 if worker's race is white = 0 otherwise

age = workers age in years

fatrate = worker's risk of fatality measured in terms of deaths per 100,000 workers.

the code I have so far (to estimate a linear hedonic wage equation for wages and log wages) is:

data final;

set 'final';

lwage= log(wage);

age2 = age*age;

run;

proc means data=final;

run;

proc reg data=final outest=est tableout;

model wage = male white age age2 fatrate;

model lwage= male white age age2 fatrate;

run;

proc print data=est;

title 'OUTEST data from PROC REG';

run;

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

Posted in reply to graingerad12

04-24-2016 01:13 PM

Can you translate that to a programming problem? I'm not sure I understand the question, even having worked as an actuary

If your previous question is answered, please mark it as solved.

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

Posted in reply to graingerad12

04-24-2016 01:13 PM

Can you translate that to a programming problem? I'm not sure I understand the question, even having worked as an actuary

If your previous question is answered, please mark it as solved.

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

Posted in reply to graingerad12

04-24-2016 03:37 PM

Naming variables in a MODEL statement is not enough to specify the model. You must also specify which interactions are possible. For example, your model assumes that the effect of being white is the same for males and females (no *male* and *white* interaction) on paid wage.

There is also cause to worry about the definition of variable *fatrate*. Does it reflect objective fatality rate or the rate perceived by the worker?

PG