Lance Bachmeier

NEW EVIDENCE ON ASYMMETRIC GASOLINE PRICE RESPONSES

In a 1997 paper, Borenstein, Cameron, and Gilbert (BCG) claim that gasoline prices rise quickly following an increase in the price of crude oil, but fall slowly following a decrease. This note estimates an error-correction model with daily spot gasoline and crude-oil price data over the period 19851998 and finds no evidence of asymmetry in wholesale gasoline prices. The sources of the difference in results are twofold. First, we use the standard Engle-Granger two-step estimation procedure, whereas BCG used a nonstandard estimation methodology. Second, even using BCGs nonstandard specification, the use of daily rather than weekly data yields little evidence of price asymmetry.

MONEY GROWTH AND INFLATION IN THE UNITED STATES

Specification tests reject a linear inflation forecasting model over the period 1959–2002. Based on this finding, we evaluate the out-of-sample inflation forecasts of a fully nonparametric model for 1994–2002. Our two main results are that: (i) nonlinear models produce much better forecasts than linear models, and (ii) including money growth in the nonparametric model yields marginal improvements, but including velocity reduces the mean squared forecast error by as much as 40%. A threshold model fits the data well over the full sample, offering an interpretation of our findings. We conclude that it is important to account for both nonlinearity and the behavior of monetary aggregates when forecasting inflation.

Is the term structure nonlinear? A semiparametric investigation

A semiparametric error correction model (ECM) is estimated using US term structure data. We use 5 and 10 year interest rates to predict short-term (1 month to 12 month) interest rates. It is found that the semiparametric ECM model predicts better than the popular linear ECM. These results provide further evidence of nonlinearity in the term structure.

Identification in models of gasoline pricing

This paper presents evidence that the price of oil does not respond contemporaneously to shocks to the US gasoline market. We find no support for the hypothesis of feedback from the US gasoline market to the price of oil, justifying the identification of impulse response functions by applying a Cholesky decomposition (see, e.g., Kilian (2010)). Our results have implications for tests of asymmetric gasoline price responses and forecasting models of the price of crude oil.

Do jet fuel price movements help forecast airline fares and the demand for air travel

ABSTRACT The paper studies the predictive content of jet fuel prices for the U.S. aviation industry through in-sample and out-of-sample forecasting exercises. Our results suggest the possibility of limited improvements in the predictions of airline fares, and little evidence of predictability from jet fuel prices to measures of air travel demand.