ABSTRACT
COVID-19, a respiratory virus violently spread worldwide, has deeply affected people’s daily life and travel behaviors. We adopted an autoregressive distributed lag model to analyze changes in travel patterns in Houston, Texas during COVID-19. The results indicated that visit patterns and changes in COVID-19 cases a week prior heavily influence the following week’s behaviors. Additionally, unemployment claims, median minimum dwell time, and workplace visit activity played a major role in predicting total foot traffic. Notably, transit systems have seen an overall decrease in usage but were not significant in estimating total foot traffic. This model showcased a unique method of quantifying and analyzing travel behaviors in Houston in response to COVID-19.
5. Acknowledgements
Thank you to Safegraph, Google, the Federal Reserve Bank of Dallas, the Texas Workforce Commission, and the Texas State Department for accurate documentation of data used in our analysis. Thank you to Nathaniel Degen for collecting Safegraph Mobility Data within Houston and to Shunhua Bai for guidance in the design of the final model. The authors confirm contribution to the paper as follows. The construct, other data collections, and derived models were performed by Mira Bhat. Interpretation, analysis, and preparation of the manuscript draft were conducted by Mira Bhat with assistance from Dr. Jiao and Dr. Azimian. This work was supported by University of Texas good system grand challenge and USDOT CM2 University Transportation Center at University of Texas Austin.
Declarations of interest
No potential conflict of interest was reported by the authors.
Disclosure statement
No potential conflict of interest was reported by the authors.