Dawit Zerom

Empirical Analysis of Ambulance Travel Times: The Case of Calgary Emergency Medical Services

Using administrative data for high-priority calls in Calgary, Alberta, we estimate how ambulance travel times depend on distance. We find that a logarithmic transformation produces symmetric travel-time distributions with heavier tails than those of a normal distribution. Guided by nonparametric estimates of the median and coefficient of variation, we demonstrate that a previously proposed model for mean fire engine travel times is a valid and useful description of median ambulance travel times. We propose a new specification for the coefficient of variation, which decreases with distance. We illustrate how the resulting travel-time distribution model can be used to create probability-of-coverage maps for diagnosis and improvement of system performance.

On Additive Conditional Quantiles With High-Dimensional Covariates

We investigate the estimation of the conditional quantile when many covariates are involved. In particular, we model the conditional quantile of a response as a nonlinear additive function of relevant covariates. For this setup, we propose a nonparametric smoother to estimate the unknown functions. The estimator provides direct computation of the nonlinear functions. Because it does not require any iteration, the estimator allows fast and routine data analysis. On the theoretical front, we also show asymptotic properties of the estimator, including mean squared error and limiting distribution. The theory confirms that for moderate dimension of the covariates, the estimator escapes the “curse of dimensionality” problem. Both simulated and real data examples are provided to illustrate the methodology.

A Semiparametric Analysis of Gasoline Demand in the United States Reexamining The Impact of Price

The evaluation of the impact of an increase in gasoline tax on demand relies crucially on the estimate of the price elasticity. This article presents an extended application of the Partially Linear Additive Model (PLAM) to the analysis of gasoline demand using a panel of U.S. households, focusing mainly on the estimation of the price elasticity. Unlike previous semiparametric studies that use household-level data, we work with vehicle-level data within households that can potentially add richer details to the price variable. Both households and vehicles data are obtained from the Residential Transportation Energy Consumption Survey (RTECS) of 1991 and 1994, conducted by the U.S. Energy Information Administration (EIA). As expected, the derived vehicle-based gasoline price has significant dispersion across the country and across grades of gasoline. By using a PLAM specification for gasoline demand, we obtain a measure of gasoline price elasticity that circumvents the implausible price effects reported in earlier studies. In particular, our results show the price elasticity ranges between −0.2, at low prices, and −0.5, at high prices, suggesting that households might respond differently to price changes depending on the level of price. In addition, we estimate separately the model to households that buy only regular gasoline and those that buy also midgrade/premium gasoline. The results show that the price elasticities for these groups are increasing in price and that regular households are more price sensitive compared to nonregular.

Kernel-based multistep-ahead predictions of the US short-term interest rate

This paper presents a comparison of prediction performances of threekernel-based nonparametric methods applied to the U.S. weekly T-bill rate.Predictions are generated through the rolling approachfor the out-of-sample period 1989-1993. We compare the multistep-aheadprediction performance of the conditional mean, the conditionalmedian, and the conditional mode with the performanceof the benchmark random walk model. Using four predictionevaluation criteria, it is shown that two of the threepredictors are superior -- or at least equal --to the random walk at prediction horizons 1 - 5.In addition, by combining two of the three predictors, a significantimprovement in prediction accuracy is obtainedat all prediction horizons. Also the combined predictions resultin substantial improvement at predicting the direction of change.Further, we propose two prediction intervals based on the estimatednonparametric conditional distributionfunction. These intervals are useful when the predictivedistribution underlying the time series process is asymmetricor multi-modal. Finally, we assess the choice of the bandwidthin the kernel-based prediction methods through a recently proposedmethod for evaluating the estimated prediction densities.

A Multivariate Quantile Predictor

ABSTRACT We introduce a nonparametric quantile predictor for multivariate time series via generalizing the well-known univariate conditional quantile into a multivariate setting for dependent data. Applying the multivariate predictor to predicting tail conditional quantiles from foreign exchange daily returns, it is observed that the accuracy of extreme tail quantile predictions can be greatly improved by incorporating interdependence between the returns in a bivariate framework. As a special application of the multivariate quantile predictor, we also introduce a so-called joint-horizon quantile predictor that is used to produce multi-step quantile predictions in one-go from univariate time series realizations. A simulation example is discussed to illustrate the relevance of the joint-horizon approach.

Cost Pass-Through in the U.S. Coffee Industry

A rich data set of coffee prices and costs was used to determine to what extent changes in commodity costs affect manufacturer and retail prices. On average, a 10-cent increase in the cost of a pound of green coffee beans in a given quarter results in a 2-cent increase in manufacturer and retail prices in that quarter. If a cost change persists for several quarters, it will be incorporated into manufacturer prices approximately cent-forcent with the commodity-cost change. Given the substantial fixed costs and markups involved in coffee manufacturing, this translates into about a 3-percent change in retail prices for a 10-percent change in commodity prices. We do not find robust evidence that coffee prices respond more to increases than to decreases in costs.

Are macroeconomic variables useful for forecasting the distribution of U.S. inflation

Much of the inflation forecasting literature examines the ability of macroeconomic indicators to predict the mean inflation accurately. For the period after 1984, the existing empirical evidence largely suggests that the likelihood of predicting inflation accurately using macroeconomic indicators is no better than a random walk model. We expand the scope of inflation predictability by exploring whether macroeconomic indicators are useful in predicting the distribution of inflation. We consider six commonly-used macro indicators and core/non-core versions of the Consumer Price Index (CPI) and the Personal Consumption Expenditure (PCE) deflator as measures of inflation. Based on monthly data and for the forecast period after 1984, we find that some of the macro indicators, such as the unemployment rate, housing starts and the term spread, provide significant out-of-sample predictability for the distribution of core inflation. An analysis of the quantiles of the predictive distribution reveals interesting patterns which would otherwise be ignored by existing inflation forecasting approaches that rely only on forecasting the mean. We also illustrate the importance of inflation distribution forecasting in evaluating some events which are of policy interest by focusing on predicting the likelihood of deflation.

A Semiparametric Analysis of Gasoline Demand in the US: Reexamining The Impact of Price

The evaluation of the impact of an increase in gasoline tax on demand relies crucially on the estimate of the price elasticity. This paper presents an extended application of the Partially Linear Additive Model (PLAM) to the analysis of gasoline demand using a panel of US households, focusing mainly on the estimation of the price elasticity. Unlike previous semi-parametric studies that use household-level data, we work with vehicle-level data within households that can potentially add richer details to the price variable. Both households and vehicles data are obtained from the Residential Transportation Energy Consumption Survey (RTECS) of 1991 and 1994, conducted by the US Energy Information Administration (EIA). As expected, the derived vehicle-based gasoline price has significant dispersion across the country and across grades of gasoline. By using a PLAM specification for gasoline demand, we obtain a measure of gasoline price elasticity that circumvents the implausible price effects reported in earlier studies. In particular, our results show the price elasticity ranges between −0.2, at low prices, and −0.5, at high prices, suggesting that households might respond differently to price changes depending on the level of price. In addition, we estimate separately the model to households that buy only regular gasoline and those that buy also midgrade/premium gasoline. The results show that the price elasticities for these groups are increasing in price and that regular households are more price sensitive compared to non-regular.

Mean squared error properties of the kernel-based multi-stage median predictor for time series

We propose a kernel-based multi-stage conditional median predictor for [alpha]-mixing time series of Markovian structure. Mean squared error properties of single-stage and multi-stage conditional medians are derived and discussed.

Multi-stage kernel-based conditional quantile prediction in time series

We present a multi-stage conditional quantile predictor for time series of Markovian structure. It is proved that at any quantile level, p ∈ (0, 1), the asymptotic mean squared error (MSE) of the new predictor is smaller than the single-stage conditional quantile predictor. A simulation study confirms this result in a small sample situation. Because the improvement by the proposed predictor increases for quantiles at the tails of the conditional distribution function, the multi-stage predictor can be used to compute better predictive intervals with smaller variability. Applying this predictor to the changes in the U.S. short-term interest rate, rather smooth out-of-sample predictive intervals are obtained.