Note: All programs are
distributed freely for non-profit academic purposes only. For other uses,
please contact Pierre Perron at perron@bu.edu.
A lot of effort has been put to construct these programs and we would
appreciate that you acknowledge using a particular program in your research and
cite the relevant papers on which it is based and the author of the code.

**“Testing Jointly for Structural Changes in
the Error Variance and Coefficients of a Linear Regression Model”
(developed by Maria Dolores Gadea).**

**This zipped folder
contains matlab code to perform the procedures discussed
in the paper "Testing Jointly for Structural Changes in the Error Variance
and Coefficients of a Linear Regression Model,” (with Jing Zhou). Special
thanks to Maria (Lola). **

**“Testing for Flexible Nonlinear Trends with an Integrated or Stationary
Noise Component****” ** Gauss code version (developed by Tomoyoshi Yabu).

**“Testing for Flexible Nonlinear Trends with an Integrated or Stationary
Noise Component****” ** Matlab code
version (developed by Mototsugu Shintani).

These zipped folders contains
Gauss or Matlab codes to perform the procedures
discussed in the paper “Testing for
Flexible Nonlinear Trends with an Integrated or Stationary Noise Component”
by Pierre Perron, Mototsugu Shintani and Tomoyoshi Yabu.

**“Residuals based Tests for Cointegration
using GLS Detrending Data****” (developed by Gabriel Rodríguez and Miguel Ataurima).**

This zipped folder contains
Gauss and Matlab codes to perform the procedures
discussed in the paper "Residuals based Tests for Cointegration
using GLS Detrending Data" (with Gabriel** Rodríguez**),
Econometrics Journal 19 (2016), 84-111.

"Wald Tests for Detecting Multiple Structural Changes in Persistence" (developed by Mohitosh Kejriwal).

This zipped folder contains Gauss codes to perform the procedures discussed in the paper "Wald Tests for Detecting Multiple Structural Changes in Persistence," (with Mohitosh Kejriwal and Jing Zhou), Econometric Theory 29 (2013), 289-323.

"Estimating and Testing Multiple Structural Changes in Linear Models Using Band Spectral Regressions" (developed by Yohei Yamamoto).

This zipped folder contains Matlab codes to perform the procedures discussed in the paper "Estimating and Testing Multiple Structural Changes in Linear Models Using Band Spectral Regressions" (with Yohei Yamamoto), Econometrics Journal 16 (2013), 400-429.

"Let's Take a Break: Trends and Cycles in U.S. Real GDP", Journal of Monetary Economics 56 (2009), 749-765 (developed by Tatsuma Wada).

This zipped folder contains Matlab codes to replicate the results in the paper "Let's Take a Break: Trends and Cycles in U.S. Real GDP" (with Tatsuma Wada), Journal of Monetary Economics 56 (2009), 749-765.

**"GLS-based
Unit Root Tests with Multiple Structural Breaks both Under the Null and the
Alternative Hypotheses" (developed by Josep Lluís Carrion-i-Silvestre).**

**This Gauss code is a companion to the paper
GLS-based Unit Root Tests with Multiple Structural Breaks both Under the Null
and the Alternative Hypotheses" (Econometric Theory, 25 (2009), 1754-1792.**

**"Unit Root
Tests Allowing for a Break in the Trend Function Under
Both the Null and Alternative Hypotheses" (developed by Dukpa Kim).**

This
Matlab code is a companion to the paper **"Unit Root Tests Allowing for a Break in the
Trend Function Under Both the Null and Alternative
Hypotheses"** **(Journal
of Econometrics 148, 2009, 1-13). **

Estimating Deterministic Trends with an Integrated or Stationary Noise Component (developed by Tomoyoshi Yabu). Revised March 2009. The Matlab version kindly developed and provided by Lola Gadea (2017) is available here.

This GAUSS code is a companion to the paper Estimating Deterministic Trends with an Integrated or Stationary Noise Component (Journal of Econometrics, 151, 2009, 56-69.). It contains a procedure to compute the test statistic and confidence intervals for the slope of a trend function that are valid whether the noise component is stationary or integrated.

Testing for Shifts in Trend with an Integrated or Stationary Noise Component, (developed by Tomoyoshi Yabu), Revised March 2009. The Matlab version kindly developed and provided by Lola Gadea (2017) is available here.

This GAUSS code is a companion to the paper Testing for Shifts in Trend with an Integrated or Stationary Noise Component, (with Tomoyoshi Yabu), Journal of Business and Economic Statistics 27 (2009), 369-396. It compute the test for breaks in the trend function of a time series valid whether the noise component is stationary or integrated.

Estimating and testing structural changes in multivariate regressions (Econometrica, 2007) (developed by Zhongjun Qu). Revised January 2007.

This GAUSS code is a companion to the paper Estimating and testing structural changes in multivariate regressions. It contains procedures to do the following for a multi-equations model that allows multiple structural changes and arbitrary restrictions on the coefficients: 1) Estimate the model and construct confidence intervals for the estimates (break dates and coefficients); 2) Compute various tests for the presence of breaks; 3) Estimate and construct confidence intervals for the break dates of a two equations locally ordered break model (and construct the tests for the presence of breaks).

Estimating restricted structural change models (developed by Zhongjun Qu). Revised October 2004.

This GAUSS code is a companion to the paper Estimating Restricted Structural Change Models. It contains procedures to do the following for a single equation model that allows multiple structural changes and arbitrary restrictions on the coefficients: 1) Estimate the model and construct confidence intervals for the estimates (break dates and coefficients); 2) Compute the sup-F test for breaks with restrictions; 3) Simulate the critical values of the restricted structrual change sup-F test.

Computation
and hypothesis testing in models with multiple structural changes* *(developed by Pierre Perron). Revised
2004.

This
GAUSS code contains an extensive program that allows: constructing estimates of
parameters in models with multiple stuctural change
(the main ingredient being a dynamic programming algorithm); estimating the
number of breaks (using information criteria or sequential hypothesis testing);
constructing confidence intervals (in particular for the estimated break
dates); testing for structural changes (various methods including global and
sequential). The program allows one to estimate pure as well as partial structural
change models. It also provide options to allow for
heterogeneity and/or serail correlation in the data
and the errors across segments. The code is described in more detail in
"Computation and Analysis of Multiple Structural Change Models,"
(with Jushan
Bai), J*ournal of Applied
Econometrics** *18 (2003), 1-22*.*

*A Matlab version is
available here
(developed by Yohei Yamamoto; June 2012; revised May 2018).*

Unit root tests
with a one time structural change* *(developed
by Serena Ng and Pierre Perron)

This zip file contains RATS procedures which test for a unit root allowing for a structural break when the time of the break is unknown. Each procedure allows one to use any of five different ways to select the lag order in an augmented autoregression over a prespecified range from 0 to kmax. Four models are considered; Model 0: change in mean with non-trending data; Model 1: change in level with trending data; Model 2: change in level and slope of the trend; Model 3: change in the slope of the trend with the segments joined at the time of break There are many diffrent procedures, in particular, depending on the way the break point is selected and how the data are demeaned or detrended. Two sets of procedures are included one for Rats Version 3.1 and one for Version 4. Also included is a directory called pval which contains a small dos program that allows retrieving the asymptotic p-value of a given estimated test statistic (developed by Timothy J. Vogelsang).

Unit root tests with GLS detrended data and the MIC to select the autoregressive order (developed by Serena Ng)

This is a Gauss code that construct a variety of unit root tests (M-tests,
Point optimal tests and ADF tests) using GLS-detrended
data. To select the order of the autoregression, an
option allows using the MIC developed in "Lag Length Selection and the
Construction of Unit Root Tests With Good Size and
Power,'' (with Serena Ng), *Econometrica**, *69 (2001), 1519-1554*.*