摘要：Liquidity costs in futures markets are not observed directly because bids and offers occur in an open outcry pit and are not recorded. Traditional estimation of these costs has focused on bidask spreads using transaction prices. However, the bid-ask spread only captures the tightness of the market price. As the volume increases measures of market depth which identify how the order flow moves prices become important information. We estimate market depth for lean hogs and live cattle markets using a Bayesian MCMC method to estimate unobserved data. While the markets are highly liquid, our results show that cost- and risk-reducing strategies may exist. Liquidity costs are highest when larger volumes are traded at distant contracts. For hogs the market becomes less liquid prior to the expiration month. For cattle this occurs during the expiration month when the liquidity risk is also higher. For both markets this coincides with periods of low volume. For the nearby contract highest trading volume occurs at the beginning of the month prior to expiration and lowest trading volume occurs in the expiration month. For both commodities the cumulative effect of volume on price change may lead to liquidity costs higher than a tick.
