PROC SURVEYREG: ESTIMATE Statement :: SAS/STAT(R) 9.2 User's Guide, Second Edition

The SURVEYREG Procedure

ESTIMATE Statement

ESTIMATE ’label’ effect values </ options> ;

ESTIMATE ’label’ effect values < ... effect values> </ options> ;

You can use an ESTIMATE statement to estimate a linear function of the regression parameters by multiplying a row vector $\text{[math]}$ by the parameter estimate vector $\text{[math]}$ .

Each term in the MODEL statement, called an effect, is a variable or a combination of variables. You can specify an effect with a variable name or with a special notation by using variable names and operators.

For more details about how to specify an effect, see the section Specification of Effects in Chapter 39, The GLM Procedure.

PROC SURVEYREG checks the linear function for estimability. (See the SINGULAR= option).

The procedure displays the estimate $\text{[math]}$ along with its standard error and t test. If you specify the CLPARM option in the MODEL statement, PROC SURVEYREG also displays confidence limits for the linear function. By default, the degrees of freedom for the t test equal the number of clusters (or the number of observations if there is no CLUSTER statement) minus the number of strata. Alternatively, you can specify the degrees of freedom with the DF= option in the MODEL statement.

You can specify any number of ESTIMATE statements, but they must appear after the MODEL statement.

In the ESTIMATE statement,

label

identifies the linear function $\text{[math]}$ in the output. A label is required for every function specified. Labels must be enclosed in single quotes.

effect

identifies an effect that appears in the MODEL statement. You can use the INTERCEPT keyword as an effect when an intercept is fitted in the model. You do not need to include all effects that are in the MODEL statement.

values

are constants that are elements of the vector $\text{[math]}$ associated with the effect. For example, the following statement forms an estimate that is the difference between the parameters estimated for the first and second levels of the CLASS variable A:

                     estimate 'A1 vs A2' A  1 -1;

You can specify the following options in the ESTIMATE statement after a slash (/):

DIVISOR=value

specifies a value by which to divide all coefficients so that fractional coefficients can be entered as integers.

For example, you can use

   estimate '1/3(A1+A2) - 2/3A3' a 1 1 -2 / divisor=3;

instead of

   estimate '1/3(A1+A2) - 2/3A3' a 0.33333 0.33333 -0.66667;

E

displays the entire coefficient vector $\text{[math]}$ .

NOFILL

requests no filling in higher-order effects. When you specify only certain portions of the vector $\text{[math]}$ , by default PROC SURVEYREG constructs the remaining elements from the context. (See the section Specification of ESTIMATE Expressions in Chapter 39, The GLM Procedure. ) When you specify the NOFILL option, PROC SURVEYREG does not construct the remaining portions and treats the vector $\text{[math]}$ as it is defined in the ESTIMATE statement.

SINGULAR=value

tunes the estimability checking. If $\text{[math]}$ is a vector, define ABS( $\text{[math]}$ ) to be the largest absolute value of the elements of $\text{[math]}$ . For a row vector $\text{[math]}$ , define

$\text{[math]}$

If ABS( $\text{[math]}$ ) is greater than $\text{[math]}$ *value, then $\text{[math]}$ is declared nonestimable. Here, $\text{[math]}$ is the matrix $\text{[math]}$ . The value must be between 0 and 1; the default is $\text{[math]}$ .

Top of Page