Statistical inferences for single-index models with measurement errors |
| |
Authors: | Zhensheng Huang Wen Lou |
| |
Affiliation: | School of Science, Nanjing University of Science and Technology, Nanjing, People''s Republic of China |
| |
Abstract: | An important factor in house prices is its location. However, measurement errors arise frequently in the process of observing variables such as the latitude and longitude of the house. The single-index models with measurement errors are used to study the relationship between house location and house price. We obtain the estimators by a SIMEX method based on the local linear method and the estimating equation. To test the significance of the index coefficient and the linearity of the link function, we establish the generalized likelihood ratio (GLR) tests for the models. We demonstrate that the asymptotic null distributions of the established GLR tests follow -distributions which are independent of nuisance parameters or functions. Finally, two simulated examples and a real estate valuation data set are given to illustrate the effect of GLR tests. |
| |
Keywords: | Single-index model measurement errors local linear regression SIMEX generalized likelihood ratio test |
|
|