I have a real GDP data base 1985 price from 1988 to 2009 and real GDP data base 2000 from 2009 to 2017, plus their nominal equivalent. How can I rebase the former half to 2000 prices?
I have log transform my model already, choose per capita variables but multicollinearity is still a problem.
I plan to drop the variable which has the highest vif (over 200)... however, I read somewhere that I will be subjected to specification error/bias.
I need to address this one since...