I am trying to explore the relationship between Remittance and GDP in Bangladesh perspective. But, I am facing problem with GDP data as it had multiple base over the periods from 1972.
Will it make any problem if I run this data against remittance in regression [ GDP = f (remittance)] ?
Is there anyway to make this GDP data uniform?
Will a dummy variable for the base year solve this problem?
The dummy variable will tell you if the breaks occationed by the base years are significant or not. You should express the data on a common base year, assuming that this is not done by the offical source. The attach little spreadsheet shows how to change the base for say CPI.
I guess yes. You need to first check the seasonality of remittance data, and de-seasonalize if required (X11 or X12) and then also ensure entire GDP series is on the same base and then proceed further.
hope this helps!!
GDP is usually expressed in different ways. Adjuasated to cost of living, real, in absolute terms, as index numbers showing the change over years, etc.Usually the central / reserve bank in its annual reports releases figures that have been derived through the same method. It should be possible to create a time series with numbers that are comparable. I am not familiar to Bangladesh, but in INdia the Economic Survey, which is released before the budget for the following financial year is announced hs figures that can be used for time series.... and one can create index numbers that refer to the same base year; if only dat ais available that is already in indices which refer to different base years the whole becomes more tricky, but I doubt that this is the case. One could resort to date published by World Bank or IMF, which have databses that go several decades back into the past.
- Badges
- Science topic
- Similar topics
- Publishing