Detecting and correcting for heteroscedasticity / Teoh Sian Hoon
The discussion of this paper focuses on detecting the presence of heteroscedasticity using two methods simultaneously: (1) a graph of the residuals versus the independent variable, and (2) Goldfeld- Quandt test. These two methods were applied for the accuracy of detecting heteroscedasticity. Weighte...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universiti Teknologi MARA, Pulau Pinang & Pusat Penerbitan Universiti (UPENA)
2007
|
| Subjects: | |
| Online Access: | http://ir.uitm.edu.my/id/eprint/14876/ http://ir.uitm.edu.my/id/eprint/14876/1/AJ_TEOH%20SIAN%20HOON%20ESTEEM%2007.pdf |
| Summary: | The discussion of this paper focuses on detecting the presence of heteroscedasticity using two methods simultaneously: (1) a graph of the residuals versus the independent variable, and (2) Goldfeld- Quandt test. These two methods were applied for the accuracy of detecting heteroscedasticity. Weighted least squares was used in
correcting the presence of heteroscedasticity. A clear picture on detecting and correcting heteroscedasticity was discussed in a case study. SPSS was used in plotting the graph of residuals against the independent variable and also in analyzing data using weighted least squares. |
|---|