In this recipe, we investigate how to make a non-stationary time series stationary by using the following transformations:
- Deflation: Accounting for inflation in monetary series using the Consumer Price Index (CPI)
- Natural logarithm: Making the exponential trend closer to linear
- Differencing: Taking the difference between the current observation and a lagged value (observation x time points before it)
We use the same data that we used in the Testing for stationarity in time series recipe. The conclusion from that recipe was that the time series of monthly gold prices from 2000-2011 was not stationary.
