This is the code and data for the paper Hernández-Lobato J. M., Houlsby N. and Ghahramani Z. Probabilistic Matrix Factorization with Non-random Missing Data, In ICML, 2014. The folder "nipsDataset" contains the data from the reviewer bidding process for the 2013 NIPS conference, as described in the paper. The folder "paper" contains the latex code for the paper. The folder "code" contains the code for the methods MF-MNAR and MF-MAR as described in the above reference. The folder "code/data" contains the data for the dataset SMF-MNAR as described in the above reference as well. The script "code/SMF-MNAR/MF-MAR/simulation1/runExperiment.sh" runs the method MF-MAR in the first train-test partition of the data in the dataset SMF-MNAR. The results of the method are stored in the folder "code/SMF-MNAR/MF_MAR/simulation1/results". The script "code/SMF-MNAR/MF-MNAR/simulation1/runExperiment.sh" runs the method MF-MNAR in the first train-test partition of the data in the dataset SMF-MNAR. The results of the method are stored in the folder "code/SMF-MNAR/MF-MNAR/simulation1/results". You may need to compile some of the c files to run from R. You can do that by running the scripts "code/SMF-MNAR/MF-MNAR/simulation1/compile.sh" and "code/SMF-MNAR/MF-MAR/simulation1/compile.sh". Before doing this, you need to install the GNU Scientific Library and replace in these files the paths "/users/jmh233/gsl-1.9/lib/" and "/users/jmh233/R-2.14.1/include" with the paths to your copy of the GNU Scientific Library and of R, respectively.