Hi. I am using mixed modeling for my multilevel data across countries, industries, and firms. I need your comment on a dilemma I have. My stata code is: "mixed DV(R&D) IVs || _all:R.industry || country: || firm: , mle variance cov(uns)

My dependent variable is R&D investments and my independent variable is performance. Both DV and IV are at the firm level. The degree of R&D in the industry level is known to be a significant predictor of R&D at the firm level. So industry R&D is mostly used as a control in the other papers. My question is that since I am using industry random intercept, should I still use R&D in the industry level as a control? Or it is not necessary?