Python - Pyspark

The purpose of this exercise was to predict housing prices for a certain year based on prices the year before. A high-quality predictive model will have an RMSE of .2 to .5. The RMSE for this model was over 140,000, making it a useless tool in predicting housing prices. The Real Estate market deals with a number of variables that effect housing prices, picking one, recent history, will not paint a clear picture. If the model was only going to look at yearly prices, then the model would need to go back decades to help identify trends, but even that would not be accurate when predicting year by year pricing.
In order to more accurately predict yearly pricing, other factors would have to be considered. Local population growth would indicate how supply and demand would effect the market. This would need to be partnered with demographic studies showing who is moving to an area, or who is moving out. Interest rates would play a big part in how much home a person could buy, and the overall economy is an important variable. In short: the model needs more variables.