AUTHORS: Farrokh Alemi, Timothy P. Coffin
Download as PDF
ABSTRACT: At its core, performance improvement requires a causal inference. The alternative causes of poor performance needs to be identified and statistically controlled so that the effect of the new intervention on performance can be assessed. Unfortunately, current control charts are not based on principles of causal inference. Objective: To provide a method of assessing causal impact of an intervention while controlling for alternative explanations. Methods: The impact of the intervention (cases) is compared to a counterfactual, simulated, control. The data are stratified by combination of alternative causes. Within each stratum cases after the intervention are compared to weighted controls, where weights are chosen so that the frequency of alternative explanations among cases and controls are the same. The methodology is applied to changes in stock prices after election of President Trump, with general trend in the economy and general trend in the healthcare stock prices being the alternative explanation. The impact of the election is examined after removing the effects of alternative explanations. Results: Impact of election on stock prices differs after we control for alternative explanations for rise of stock prices. Conclusions: Causal control charts may be useful in situations where several competing causes exists for changes in performance
KEYWORDS: Causal analysis, Causal control charts, Counterfactuals, Stratification.
REFERENCES:
[1] I. Bass, Six Sigma Statistics with excel and Minitab, 1st ed. McGraw-Hill Professional Publishing, 2007.
[2] J. Pearl, “Causal inference in statistics: An overview,” Stat. Surv, pp. 96–146, 2009.
[3] S. Morgan and C. Winship, Counterfactual and Causal Inference. Cambridge University Press, 2007.
[4] F. Alemi and D. Oliver, “Tutorial on riskadjusted P-charts.,” vol. 10, no. 1, pp. 1– 9, Fall 2001.
[5] J. Cockings, D. Cook, and R. Iqbal, “Process monitoring in intensive care with the use of cumulative expected minus observed mortality and riskadjusted P charts.,” Crit Care, vol. 10, no. 1, p. R28, Feb. 2006.
[6] D. Cook, M. Coory, and R. Webster, “Exponentially weighted moving average charts to compare observed and expected values for monitoring risk-adjusted hospital indicators,” BMJ Qual Saf, vol. 20, no. 6, pp. 469–74, Jun2011.
[7] D. Cook, G. Duke, G. Hart, and K. Lee, “Review of the application of riskadjusted charts to analyse mortality outcomes in critical care,” Crit Care Resusc, vol. 10, no. 3, pp. 239–51, Sep. 2008.
[8] M. Hart, J. Robertson, R. Hart, and K. Lee, “Application of variables control charts to risk-adjusted time-ordered healthcare data,” Qual Manag Health Care, vol. 13, no. 2, p. 13(2):99-119., Jun. 2004.
[9] D. B. Rubin, “Estimating causal effects of treatments in randomized and nonrandomized studies.,” J. Educ. Psychol., vol. 66, no. 5, pp. 688–701, 1974.
[10] D. B. Rubin, “Assignment to Treatment Group on the Basis of a Covariate,” J. Educ. Stat., vol. 2, no. 1, pp. 1–26, Mar. 1977.
[11] D. B. Rubin, “Bayesian Inference for Causal Effects: The Role of Randomization,” Ann. Stat., vol. 6, no. 1, pp. 34–58, 1978.
[12] H. Wong and D. B. Rubin, “Causal Inference Using Potential Outcomes: Design, Modeling, Decisions,” Am. Stat., vol. 62, no. 3, pp. 275–278, 2008.
[13] H. Wong and D. B. Rubin, “Rubin, D. B. (2005), ‘Causal Inference Using Potential Outcomes: Design, Modeling, Decisions,’ ‘Journal Of The American Statistical Association,’ 100, 322-331: Comment by Wong and Reply,” Am. Stat., vol. 62, no. 3, pp. 275–278, 2008.
[14] J. J. Heckman, “Sample Selection Bias as a Specification Error,” Econometrica, vol. 47, no. 1, pp. 153–161, 1979.
[15] J. J. Heckman, “Econometric Causality,” Int. Stat. Rev. Rev. Int. Stat., vol. 76, no. 1, pp. 1–27, 2008.
[16] L. A. Paul, E. J. Hall, and J. D. Collins, Causation and Counterfactuals. Cambridge, Mass: A Bradford Book, 2004.
[17] F. Alemi and D. H. Gustafson, Decision Analysis for healthcare managers. Chicago: Health Administration Press, 2007.
WSEAS Transactions on Business and Economics, ISSN / E-ISSN: 1109-9526 / 2224-2899, Volume 15, 2018, Art. #25, pp. 259-272
Copyright © 2018 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0