I am new to R and I'm trying to do a time-series analysis of Dogecoin and google searches (google trends) for dogecoin.
I have noticed, that when I use the gtrends package, that when I specify a timeframe that is longer than 90 days, I will only get the weekly average searches. moreover, I do not get the specific amount of searches, but the relative search volume ranking von 0-100 (hits) within this time period (the day with the most amount of searches always gets 100).
This means that I cannot simply get multiple 90 day periods and stitch them together since I'm given the relative amount of searches for each period.
I suspect there must be a way to get overlapping time periods and then adjust the values accordingly. Even though this seems to be a quite obvious problem, I was unable to find a solution online. It should be possible to write a function for this, right?
I have no clue on how to do this with regards to the math and syntax for this endeavor. could someone help me out?
Asked
Active
Viewed 730 times
2

Marcus
- 105
- 1
- 2
- 8