Open Access Open Access  Restricted Access Subscription or Fee Access

Application of an Ordinal Probit Regression Model in predicting emergency response rates in the Fire Department of New York City

Ivan L. Pitt, PhD

Abstract


This article introduces the use of an Ordinal Probit Regression Model to predict emergency response rates in the Fire Department of New York City (FDNY) when data are of the count-type variety and ordinal. The main objective of this article is to model the effects of boroughs, emergency incident types, and the volume of emergency incidents (counts) on response rates for the years 2010-2016. The model framework discusses the model selection criteria when the proportional odds assumptions for ordinal models are no longer valid, and a model in which scale effects are allowed to vary among emergency incident types is preferred. This statistical insight can be used by elected officials and city agencies to evaluate FDNY’s emergency response availability, capability, and operational performance in order to improve emergency services in the five boroughs.


Keywords


ordinal probit regression, NYC FDNY, emergency response rates, EMS

Full Text:

PDF

References


Moore-Merrell L: Understanding and Measuring Fire Department Response Times. Frisco, TX: Lexipol, 2019. Available at https://www.lexipol.com/resources/blog/understanding-and-measuringfire-department-response-times/. Accessed November 20, 2019.

Blackwell T, Kaufman J: Response time effectiveness: Comparison of response time and survival in an urban emergency medical services system. Acad Emerg Med. 2002; 9(4): 288-295.

Neil NC: The relationships between fire service response time and fire outcomes. Fire Technol. 2009; 46(3): 665-676.

Blanchard I, Doig CJ, Hagel BE, et al.: Emergency medical services response time and mortality in an urban setting. Prehosp Emerg Care. 2012; 16(1): 142-151.

Wilde E: Do emergency medical system response times matter for health outcomes? Health Econ. 2012; 22(7): 790-806.

Matteson DS, Mclean MW, Woodard DB, et al.: Forecasting emergency medical service call arrival rates. Ann Appl Stat. 2011; 5(2B): 1379-1406.

Zhou Z: Predicting ambulance demand: Challenges and methods. In ICML Workshop on Data4Good: Machine Learning in Social Good. 2016: 11-15. Available at https:arxiv.org.pdf. Accessed November 24, 2019.

Henderson SG: Operations research tools for addressing current challenges in emergency medical services. In Cochran JJ, Cox LA, Keskinocak P, Kharoufeh JP, Smith JC (eds): Wiley Encyclopedia of Operations Research and Management Science. Hoboken: Wiley, 2009.

Fraley C, Raftery AE: Bayesian regularization for normal mixture estimation and model-based clustering. J Classif. 2007; 24: 155-181.

Scrucca L, Fop M, Murphy TB, et al.: mclust 5: Clustering, classification and density estimation using Gaussian finite mixture models. R J. 2016; 8(1): 205-233.

Law AM, Kelton DW: Simulation Modeling and Analysis. 3rd ed. New York: McGraw-Hill, 2000.

Upson R, Notarianni K: Quantitative evaluation of fire and EMS mobilization times. Fire Protection Research Foundation. Final Report. 2010.

Pitt IL: Graphical analysis of an ordinal probit regression model in predicting emergency response rates in the Fire Department of New York City. Working Paper. 2020.

Fernandez-Val I: Fixed effects estimation of structural parameters and marginal effects in panel probit models. J Economet. 2009; 150: 71-85.

Cameron AC, Trivedi PK: Count panel data. In Baltagi B (ed): Oxford Handbook of Panel Data. Oxford: Oxford University Press, 2014.

Agresti A: An Introduction to Categorical Data Analysis. 2nd ed. Hoboken, NJ: John Wiley & Sons, 2007.

Baltagi B: Oxford Handbook of Panel Data. Oxford: Oxford University Press, 2014.

Greene W: Econometric Analysis. 7th ed. London: Pearson, 2012.

McCullagh P: Regression models for ordinal data. J R Stat Soc Ser B. 1980; 42(2): 109-142.

Greene W, Hensher D: Modeling Ordered Choices: A Primer. Cambridge: Cambridge University Press, 2010.

Christensen RHB: Ordinal—Regression models for ordinal data. R Package Version 2019.4-25, 2019. Available at http://www.cran.r-project.org/package=ordinal/. Accessed May 23, 2019.

White H: A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica. 1980; 48(4): 817-838.

Arellano M: Computing robust standard errors for within group estimators. Oxford Bull Econ Stat. 1987; 49(4): 431-434.

Millo G: Robust standard error estimators for panel models: A unifying approach. J Stat Softw. 2017; 82(3): 1-27.

Albert A, Anderson J: On the existence of maximum likelihood estimates in logistic regression models. Biometrika. 1984; 71: 1-10.

Brant R: Assessing proportionality in the proportional odds model for ordinal logistic regression. Biometrics. 1990; 46(4): 1171-1178.

Elde´n L, Wittmeyer-Koch L, Nielsen HB: Introduction to Numerical Computation: Analysis and MATLAB Illustrations. Lund, Sweden: Studentlitteratur, 2004.

Allison P: Convergence failures in logistic regression. In SAS Global Forum 2008. 2008: 360.

Derr R: Predicting Inside the Dead Zone of Complete Separation in Logistic Regression. Cary, NC: SAS Institute, 2019. 3018-2019.

Pitt IL: Application of ordinal probit regression analysis in predicting economic risk and demographics in New York City. Working Paper. 2020.

NYC Mayor’s Office of Criminal Justice: Smaller safer fairer: A

roadmap to closing Rikers island. 2016. Available at https://www1.nyc.gov/assets/criminaljustice/downloads/pdfs/Smaller-Safer-Fairer.pdf. Accessed November 23, 2018.

Austin J: The proper and improper use of risk assessment in corrections. Fed. Sentencing Rep. 2004; 16(3): 1-6.

Pitt IL: The application of activity based costing for data analysis in correctional practice. Int Public Manag Rev. forthcoming. 2020.




DOI: https://doi.org/10.5055/jem.0537

Refbacks

  • There are currently no refbacks.


Copyright (c) 2022 Journal of Emergency Management