Heteroscedasticity and autocorrelation consistent covariance estimators for linear regression with equality constraints

I have a linear regression

because is not full colume rank, there is a additional equality constraint: .

I can solve above optimization problems easily and get the error matrix . Because I belive the error matrix is correlated within columns (time-lag ) , I want to estimate robust covariance (i.e. Newey West method) using the error matrix . I was wondering, is there any way to do this in MATLAB?

Answer 1

[Zuber]

Hi Mingyang,

I understand that you want to estimate the heteroscedasticity and autocorrelation consistent covariance estimator specifically using Newey West Method for linear regression. In order to obtain the covariance estimate, you can use the ‘hac’ function.

[EstCoeffCov,se,coeff] = hac(X,y)

where, ‘EstCoeffCov’ is the robust covariance matrix estimate.

For more information, please refer to the following documentation: https://www.mathworks.com/help/econ/hac.html

I hope this resolves your query.