How appropriate would be, if you perform Chow breakpoint test for all points within series (without the very beginning and very ending), then find out where test suggest possible breakpoints (e.g. there are 4-5 points) and analyze them from visual and common-sense point of view - whether it is a real structural break or not?


1 Answer 1


The test is appropriate and was suggested by Quandt (1960) right after Chow (1960) had introduced it. However the correct asymptotic distribution for computing critical values and p-values was unknown until Andrews (1993) rigorously derived the required theory. An R implementation is available in R package strucchange (Zeileis et al. 2002) based on the p-values from Hansen (1997). The package also emphasizes visualizations of the sequence of test statistics to accompany the final significance tests.

This analysis can also be extended to multiple change points along with formal breakpoint estimation (Bai and Perron 2003). See Zeileis et al. (2003) for the implementation in strucchange.

  • Andrews DWK (1993). Tests for Parameter Instability and Structural Change with Unknown Change Point, Econometrica61, 821-856.
  • Bai J, Perron P (2003). Computation and Analysis of Multiple Structural Change Models, Journal of Applied Econometrics18, 1-22.
  • Chow GC (1960). Tests of Equality Between Sets of Coefficients in Two Linear Regressions. Econometrica28(3), 591–605.
  • Hansen B (1997), Approximate Asymptotic p Values for Structural-Change Tests, Journal of Business & Economic Statistics15, 60-67.
  • Quandt RE (1960). Tests of the Hypothesis That a Linear Regression System Obeys Two Separate Regimes. Journal of the American Statistical Association55(290), 324–330.
  • Zeileis A, Leisch F, Hornik K, Kleiber C (2002). strucchange: An R Package for Testing for Structural Change in Linear Regression Models, Journal of Statistical Software7(2), 1-38. http://www.jstatsoft.org/v07/i02/
  • Zeileis A, Kleiber C, Krämer W, Hornik K (2003). Testing and Dating of Structural Changes in Practice, Computational Statistics and Data Analysis44, 109-123.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.