Hello everyone, I am about to do a (S)VAR analysis of monetary policy shocks. I wish to include two variables, among others, which are indexed in different years. I only found GDP to be indexed in 2010 (2010=100) and inflation (HCPI) to be indexed either in 2005 or 2015 (2005=100 or 2015=100).
Now I am not sure if it imposes any issues to use those variables in my analysis. Should they be indexed with respect to the same year?
Thank you in advance!
BASE YEAR DIFFERENCE does not vitiate the validity of the studies. There might be reasons why the organization or government set a different base year fr various index. The job of the researcher is to study the data, not to re-make the index whilst he/she is making up his/her data from a secondary data (indices). In this case, the two indices used different base-year, there is nothing wrong with that because each variable is independent and each index is independently constructed by whichever primary source that produced the index. As a researcher using a secondary data, make the best of what could be done with the data without any manipulation---this is part of the limitation of secondary data.
I agree with Dr. Paul Louangrath first sentence and simply taking the logarithm of varables takes care of index issues!
Splicing the data is important since many journals want a fixed base period. There is no problem with doing this. In fact, I do this in my own research but I also, have important sources from the literature to support the use of logarimthic transformation.
Thank you very much for your answers! I really appreciate it, I think I will probably go with the logarithm method and leave the data as they are as I want to stay minimalistic with the analysis. I wish you a great day!
Chuck A Arize can you please share some sources/references that support logarithmic transformation?