COPULA BASED PREDICTION MODEL
Pearson's correlation coefficient is a measure of association which widely used in many fields. However, Pearson's correlation coefficient has limitation to detect non-linear association. Kendall's Tau and Spearman's Rho are measures of association which can detect both linear an...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/16770 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Pearson's correlation coefficient is a measure of association which widely used in many fields. However, Pearson's correlation coefficient has limitation to detect non-linear association. Kendall's Tau and Spearman's Rho are measures of association which can detect both linear and non-linear association and can be stated in a multivariate distribution function model, named copula. A prediction model is established to predict a value for future observation. Generally, the model prediction is based on Pearson's correlation coefficient. However, considering the limitation that Pearson's correlation coefficient has, in this final project will be established a prediction model based on copula. |
|---|