We consider a two-stage estimation method for linear regression that uses the lasso in Tibshirani (1996) to screen variables and re-estimate the coefficients using the least-squares boosting method in Friedman (2001) on every set of selected variables. Based on the large-scale simulation experiment in Hastie et al. (2020), the performance of lassoed boosting is found to be as competitive as the relaxed lasso in Meinshausen (2007) and can yield a sparser model under certain scenarios. An application to predict equity returns also shows that lassoed boosting can give the smallest mean square prediction error among all methods under consideration.