#TITLE_ALTERNATIVE#
Kalman Filter (KF) is an algorithm that combine of the models and observations. Using data KF can to estimate some unobserved parameter such as to estimate a jump in stock return, pricing of commodity state in future time. KF is going to forecast in cases of nonlinier model optimally. Non linier mod...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/16927 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Kalman Filter (KF) is an algorithm that combine of the models and observations. Using data KF can to estimate some unobserved parameter such as to estimate a jump in stock return, pricing of commodity state in future time. KF is going to forecast in cases of nonlinier model optimally. Non linier model is handled by EnKF (Ensemble Kalman Filter) based on sequentially updating. In this Final Project KF is used to estimate value of the spot prices in accordance of the simulation on the Microsoft Excel while EnKF is implemented to estimate the permeability of the pressure data sequentially based on radial flow model in cases of the reservoir using some initially assumption of the distribution pressure and the permeability with Matlab. Time series model especially AR(p) is used to represents the EnKF permeability in order to refer convergence of the permeability. Calculation results of KF and EnKF give a result sufficiently accurate because it had been corrected by updating in accordance with observation. Using results of AR(p) model for the permeability use Normal and TGNU(Triangular, Gamma, Normal, Uniform) EnKF distribution assumptions give a convergence according to expectations. |
|---|