New covariance computations provide estimates for broader applications.
- Clustered and heteroscedastic-robust covariance computations.
- For stand alone use or use with olsmt.
- Compute outer product of gradients (OPG) covariance after optimization using sqpSolvemt.
// Set up olsmtControl structure
struct olsmtControl oCtl;
ctl = olsmtControlCreate();
// Set up robust SE
ctl.cov = "robust";
// Estimate model
call olsmt("detroit.sas7bdat", "homicide ~ unemployment + hourly_earn", ctl);↓
Valid cases: 13 Dependent variable: homicide Missing cases: 0 Deletion method: None Total SS: 3221.790 Degrees of freedom: 10 R-squared: 0.834 Rbar-squared: 0.801 Residual SS: 533.814 Std error of est: 7.306 F(2,10): 25.177 Probability of F: 0.000 Durbin-Watson: 1.407 Standard Prob Standardized Cor with Variable Estimate Error t-value >|t| Estimate Dep Var ---------------------------------------------------------------------------------- CONSTANT -35.9829 12.2748 -2.9315 0.015 --- --- unemployment -0.0050 0.6034 -0.0083 0.994 -0.000720 0.210142 hourly_earn 15.4872 1.9391 7.9870 0.000 0.913572 0.913406 Note: Robust standard errors reported