The Study of SpatialWeight Matrix in Space-Time Modeling Through Fuzzy Sets Approach
In this thesis determine the spatial weights matrix of Generalized-STAR models using fuzzy sets concept approach.The spatial weight was mostly formulated based on the distance closeness of the observed locations, but the fuzzy approach provides a different point of view. Fuzzy spatial weight matr...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Subjects: | |
| Online Access: | https://digilib.itb.ac.id/gdl/view/33937 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | In this thesis determine the spatial weights matrix of Generalized-STAR models
using fuzzy sets concept approach.The spatial weight was mostly formulated based
on the distance closeness of the observed locations, but the fuzzy approach provides
a different point of view. Fuzzy spatial weight matrix was built based on the value
of observation data, so the spatial locations whose observations in the same values
possible have equal influences (weight) for each other. We expected that by using
this method the heterogeneity of the spatial influences will be better accommodated.
By doing some simulations, we compared the binary, gaussian, and fuzzy spatial
weight influences on the residuals sum of squares. We observed that the fuzzy
weight gives smaller values than the binary spatial weight. In case study, the concept
of fuzzy weight matrix can be applied either. The resulting models can follow the
pattern of the actual data although still can not reach the extreme values. |
|---|