Frank C. Z. Wu, "Bayesian Collapsed Gibbs Sampling for a Stochastic Volatility Model with a Dirichlet Process Mixture", Journal of Applied Econometrics, forthcoming. All files are stored in the files wu-files.zip. Most files are ASCII files in DOS format, but the .mat files are binary. The main files are: -- vardpm_main.m: the main matlab program file to do the estimation. -- oneDPM.m: is used in main file for initialization for MCMC. -- normalWishartSV.m: is to build the class of normal-Wishart. -- randmn.m: sample random discrete number from a multinomial distribution. -- dyn_autocorr.m: function calculates autocorrelation. -- mcmc_ifac.m: function computes inefficiency factor based on a Parzen Window. -- crspvwr.csv: main data file contains observations from Jan. 2nd, 1980 to Dec. 29, 2020. Total 10,340 observations. The first column shows the date, and the second column shows the index values. The following files are used to do a comparison exercise for estimating stochastic volatility (SV). The benchmark is the algorithm in Joshua C. C. Chan, "The Stochastic Volatility in Mean Model with Time-Varying Parameters: An Application to Inflation Modeling", JBES, 35, 2017, 17-28. -- USCPI.txt: is the US quarterly CPI data file used in Chan (2017). The number of observations is 262. -- hhat_SVM.mat: contains the results using the algorithm in Chan (2017) and the above data file. -- hCI_SVM.mat: contains the results for the 90th confidence bands of hhat. We can compare the results from our algorithm with algorithm in Chan (2017). We can show that the results from our algorithm can generate very similar series. Inefficiency factors for parameters [[alpha K dlt signu2 ot]: observations from Jan. 2007 to Dec. 2020 (T2 = 3525) IF_phi = 4.3828 11.1507 7.6276 24.4610 5.5539 MH acceptance rate 12.12% observations from Jan. 1980 to Dec. 2006 (T1 = 6815) IF_phi = 5.5859 24.1862 6.7583 17.0696 4.0049 MH acceptance rate 12.75% For further information, please contact Frank C.Z. Wu