As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
The real estate price is an essential index to measure the real estate industry, urban economy, and investment policy. This paper introduces a non-parametric regression (NR)-deep learning framework, which uses the non-parametric model to predict the trend of the real estate price series, and then uses the deep learning model to capture the residual information, to achieve the effect of error correction and optimize the accuracy of real estate price prediction. The empirical results show that error correction can improve the prediction accuracy by an order of magnitude. The improvement degree of six evaluation criteria is far more than 10 times. In addition, under the error correction framework, NR-gated recurrent unit (GRU) has certain advantages in processing nonlinear complex error sequences. Compared with the SVR and LSTM model under the framework, the average improvement percentage of evaluation criteria is about 5.20% and 0.09%, and the DM statistics are all positive.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.