Time series forecast using AR-belief approach
© 2016 by the Mathematical Association of Thailand. All rights reserved. This paper aims at applying a recent new approach to predicting the growth rate of Thailand GDP. The new approach will provide uncertainty about predicted values solely from observed data without the need to supply some subject...
محفوظ في:
| المؤلفون الرئيسيون: | , , |
|---|---|
| التنسيق: | دورية |
| منشور في: |
2018
|
| الموضوعات: | |
| الوصول للمادة أونلاين: | https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85008185815&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/55978 |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| المؤسسة: | Chiang Mai University |
| الملخص: | © 2016 by the Mathematical Association of Thailand. All rights reserved. This paper aims at applying a recent new approach to predicting the growth rate of Thailand GDP. The new approach will provide uncertainty about predicted values solely from observed data without the need to supply some subjective prior distribution on unknown model parameters. This is achieved by building a belief function (i.e., a distribution of a random set) from the likelihood function given the observed data, and use it to assess prediction uncertainty. With our sampling model as an autoregressive time series model, we demonstrate em-pirically that this approach can provide a reliable con_dence interval for predicted values. |
|---|