Testing Cointegration and Mean Reversion in Currency Portfolios
Summary
The document explores whether currency prices can form stationary spreads suitable for mean-reversion analysis. It estimates a two-currency spread using ordinary least squares, then tests the residual with an augmented Dickey-Fuller procedure. It also compares a total least squares hedge ratio and applies Johansen trace and eigenvalue tests to pairs and a three-instrument portfolio that includes the inverse of a currency quote.
The reported ADF statistics for the two residual constructions do not cross the stated critical values, so these tests do not support stationarity in those examples. The Johansen output supplies candidate cointegrating relations, while a separate regression on a three-series spread is used to estimate a half-life of 53.23 periods. These are empirical diagnostics rather than evidence of a profitable strategy. The excerpt gives limited detail on data selection and robustness, and cointegration estimates can depend on model specification, sample period, and quote construction.
Key ideas
- Estimate a spread between currency prices and test its residual for stationarity before treating it as mean reverting.
- The choice of hedge-ratio estimation method can change the residual series and its test results.
- Johansen procedures can assess cointegrating relations among two or more instruments.
- A regression of spread changes on lagged spread can be used to estimate a mean-reversion half-life.
- Statistical evidence of cointegration alone does not establish trading profitability or robustness.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.