The model develops an analysis and obtains a sense of the appropriateness of simpler linear regression; correlations were calculated between the national TSA inspection checks and the full, domestic, and international T100 market data.

Passenger enplanement numbers are collected monthly and published 3-4 months after the month ends. Transportation Security Administration (TSA) checkpoint screening is closely linked to passenger enplanement. TSA data is available a day after the month has ended, making it a suitable proxy for predicting enplanement numbers.

The simple linear regression model, with seasonality, was tested for all three T100 market datasets. For each data set, seasonality was assessed from the residuals of the simple linear regression model. Using this method, it identifies a strong correlation between TSA screen data and passenger enplanements, which suggests that a regression model would be a suitable modeling methodology.

The result of the model provides preliminary estimates for passenger enplanements, with high predictability and confidence in the estimates for 3 to 12 months in advance.

Air Traffic Preliminary Estimates

Segment Preliminary Estimates

Market Preliminary Estimates

To formulate preliminary estimates, Amtrak ridership, the Bureau of Transportation Statistics (BTS) applies a predictive modeling approach. However, it specifically adapts its data inputs for the annual station ridership figures released by Amtrak.

BTS uses linear regression models that calculate correlations between preliminary indicators and full historical datasets to provide timely, "preliminary" ridership estimates before final audited numbers are available.

Amtrak Preliminary Estimates

To formulate preliminary estimates, Transit ridership, the Bureau of Transportation Statistics (BTS) applies a predictive modeling approach for major modes, including Bus, Ferry, and Rail (including heavy, light, and commuter rail). 

The model relies on Unlinked Passenger Trips (UPT) data provided by the Federal Transit Administration (FTA) via the National Transit Database (NTD); it specifically adapts its data inputs for the monthly ridership figures released by the Federal Transit Administration (FTA).

Bus Unlinked Passenger Trips Preliminary Estimates

Ferry Unlinked Passenger Trips Preliminary Estimates

Rail Unlinked Passenger Trips Preliminary Estimates

Total Unlinked Passenger Trips Preliminary Estimates

To formulate preliminary estimates of TransBorder Freight data, the Bureau of Transportation Statistics (BTS) uses a specialized predictive modeling framework that bridges the gap between raw trade figures and actionable transportation metrics between the US and Canada and the US and Mexico.

BTS applies linear regression models that correlate preliminary indicators from the FT900 U.S. International Trade in Goods and Services with detailed historical datasets. This allows for the release of preliminary monthly estimates before final audited data, which typically has a two-month lag, is fully processed.

U.S.-Canadian TransBorder Freights Preliminary Exports

U.S.-Mexican TransBorder Freight Preliminary Estimates

U.S.-North American TransBorder Freight Preliminary Estimates

U.S.-North American TransBorder Truck Freight Preliminary Estimates

U.S.-North American TransBorder Truck Imports Preliminary Estimates

U.S.-North American TransBorder Truck Exports Preliminary Estimates

To formulate preliminary estimates of Vehicle Miles Traveled (VMT), the Bureau of Transportation Statistics (BTS) and the Federal Highway Administration (FHWA) employ predictive models that integrate high-frequency monthly reports with long-term economic indicators.

The model relies on the FHWA Traffic Volume Trends (TVT) report, which is released monthly and provides estimates of VMT based on hourly traffic count data from approximately 5,000 continuous counting locations nationwide.

Vehicle Miles Traveled Preliminary Estimates