Bid-ask bounce

A Gaussian Roll model: negatively autocorrelated errors are anti-redundant evidence.

In Roll's model of the effective spread, trades bounce between bid and ask, so observed prices carry negatively autocorrelated errors around the efficient price $\mu$. Here is a Gaussian stand-in: $y_t = \mu + e_t$ with AR(1) errors $e_t = a\,e_{t-1} + \sqrt{1-a^2}\,\eta_t$ of unit marginal variance. Bounce is $a < 0$; stale, smoothed quotes are $a > 0$; $a = 0$ is the iid case.

Correlated errors couple the observations to each other, but for estimating a single $\mu$ the coupling collapses into a re-weighting of the masses: the GLS weights are the row sums of the error precision matrix, which for AR(1) come out in closed form as $\tfrac{1}{1+a}$ for the first and last trade and $\tfrac{1-a}{1+a}$ for every interior trade. Slide $a$ negative and watch every body on the rail grow — bouncing errors cancel, so the evidence is worth more than its naive iid weight and the masses more than add. Slide $a$ positive and the interior trades shrivel to near nothing: redundant evidence. The particle settles at the GLS estimate; the grey line is the naive sample mean, which ignores the bounce. Drag any trade to see the two estimators disagree.

GLS estimate naive mean total mass vs naive an interior trade is worth iid trades

The same trades, weighed twice. With $a=-0.6$ the total mass is more than triple the naive count $n$ — the posterior is far tighter than the iid analysis admits — while with $a=0.9$ twenty-four trades carry the mass of about two. Roll's estimator itself recovers the spread from exactly this signature, the negative first-order autocovariance of price changes: $s = 2\sqrt{-\mathrm{cov}(\Delta p_t, \Delta p_{t-1})}$. See Roll (1984) in the reading list, and the dictionary for where correlated residuals sit in the mechanics.