Discussions

I am trying to download some of the ATTOM data. However, I am getting some odd results when I create the subsets to meet the <50% cutoff.

For the Pre-Foreclosure data, I have three datasets that filter based on the SITUSSTATECODE: 1) Blank to ID, 2) IL to NC, and 3) ND to WY. The respective observation counts are 13,075,726; 5,147,497; and 8,906,123 for a total of 27,129,346 which is 122,874 less than the listed number of observations in the dataset (27,252,220).

The AVM also has an issue, but in this case I get MORE observations than are listed in the dataset. I get a total of 97,315,787 when the dataset lists only 97,204,353. I believe the three subsets are exhaustive and non-overlapping: 1) ESTIMATEDMAXVALUE >=450000, 2) ESTIMATEDMAXVALUE <= 300000, and 3) 300001 <= ESTIMATEDMAXVALUE <= 449999 with observation counts of 42,685,155; 31,072,622; and 23,558,010.

Is something wrong with the filtering or with the total observation count listed on the data page?