Time series
A spread follows Δsₜ = −0.05 sₜ₋₁+εₜ on daily data.
A spread follows on daily data. What is the half-life of a shock, in days?
Answer with a number in days. Fractions, powers and expressions like 23/6 or C(52,5) are read correctly in practice.
Try it
Next free question: Can three assets be pairwise correlated at −0.5?
Show the answer and worked solution
Answer
The process is an AR(1) in disguise with mean-reversion speed per day, so a shock decays as and the half-life solves , giving days. The half-life is the number every pairs trader wants, because it sets the holding period and therefore the turnover and cost budget. A useful sanity check is that a half-life of fourteen days corresponds to in the level form, which is exactly . Solving the daily recursion exactly instead, gives days, which is also accepted; the continuous formula is the one quoted in practice.
Worked solution
- Formula
- Substitute
- Solve
- Answer
Sanity check. Call it a fortnight. It sets the holding period, and therefore the turnover and the costs.
Takeaway: Half-life is divided by the mean-reversion speed.
Report an issue with this question
Learn the method
Unit roots, spurious regression and the basis of pairs trading
Of the kind reported at
More time series questions
- You regress one stock’s price on another’s and get R² = 0.95 with a t-statistic of 40.Applied · Free solution
- A spread follows an Ornstein–Uhlenbeck process dX = θ(μ−X) dt+σ dW with θ = 5 per year.Applied · Premium
- Two prices that wander together, part 2 of 3Applied · Premium
- Two stocks have a return correlation of 0.9. Does that make them a good pairs trade?Advanced · Premium
- You have five price series and suspect more than one cointegrating relationship.Advanced · Premium
- In the Engle–Granger test you regress one price on another and run a unit-root…Advanced · Premium