Log in
Language:

MERAL Myanmar Education Research and Learning Portal

  • Top
  • Universities
  • Ranking
To
lat lon distance
To

Field does not validate



Index Link

Index Tree

Please input email address.

WEKO

One fine body…

WEKO

One fine body…

Item

{"_buckets": {"deposit": "21bf05ca-21af-46e8-9bc6-cb85d69ad100"}, "_deposit": {"id": "1610", "owners": [], "pid": {"revision_id": 0, "type": "recid", "value": "1610"}, "status": "published"}, "_oai": {"id": "oai:meral.edu.mm:recid/1610", "sets": ["user-yueco"]}, "communities": ["yueco"], "control_number": "1610", "item_1583103067471": {"attribute_name": "Title", "attribute_value_mlt": [{"subitem_1551255647225": "A Simulation Study on Robust Alternatives of Least Squares Regression", "subitem_1551255648112": "en"}]}, "item_1583103085720": {"attribute_name": "Description", "attribute_value_mlt": [{"interim": "Five methods of regression namely the ordinary least squares, least absolute value, M, least\r median squares and least trimmed squares are applied to the multiple regression model. The\r several distributional assumptions of errors are considered in this study. The required data sets are\r generated by using multiple linear regression models with three explanatory variables. Then, these\r data sets are transformed into outlier contaminated data sets. After that, the performances are\r compared in terms of bias and mean squared errors criteria and then the most suitable estimation\r method is chosen. Same sets of simulated data are used and mean squared errors and bias of these\r methods are compared. It is found that ordinary least squares estimation under a heavy-tailed\r distribution does not yield outlier robust estimates. Indeed, not only with the Gaussian distribution\r but also with the skewed distributions, ordinary least squares estimators collapse in the presence of\r small levels of outlier contamination. The Huber M-estimate and bisquare M-estimate estimate\r have shown to be more appropriate alternatives to the ordinary least squares in heavy-tailed\r distributions whereas the LMS estimates are better choices for skewed data. One best method\r could not be suggested in all situations; however the use of more than one method of exploratory\r data analysis is recommended in practice."}]}, "item_1583103108160": {"attribute_name": "Keywords", "attribute_value_mlt": [{"interim": "Robust Estimators"}]}, "item_1583103120197": {"attribute_name": "Files", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_access", "date": [{"dateType": "Available", "dateValue": "2020-05-05"}], "displaytype": "preview", "download_preview_message": "", "file_order": 0, "filename": "Dr Maw Maw Khin A Simulation study on Robus Alternatives....pdf", "filesize": [{"value": "538 Kb"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 538000.0, "url": {"url": "https://meral.edu.mm/record/1610/files/Dr Maw Maw Khin A Simulation study on Robus Alternatives....pdf"}, "version_id": "fdc59110-0d33-4259-bf52-c36f5beeeb71"}]}, "item_1583103131163": {"attribute_name": "Journal articles", "attribute_value_mlt": [{"subitem_issue": "1", "subitem_journal_title": "Yangon University of Economics", "subitem_volume": "6"}]}, "item_1583103147082": {"attribute_name": "Conference papaers", "attribute_value_mlt": [{}]}, "item_1583103211336": {"attribute_name": "Books/reports/chapters", "attribute_value_mlt": [{}]}, "item_1583103233624": {"attribute_name": "Thesis/dissertations", "attribute_value_mlt": [{"subitem_supervisor(s)": []}]}, "item_1583105942107": {"attribute_name": "Authors", "attribute_value_mlt": [{"subitem_authors": [{"subitem_authors_fullname": "Maw Maw Khin, Dr."}]}]}, "item_1583108359239": {"attribute_name": "Upload type", "attribute_value_mlt": [{"interim": "Publication"}]}, "item_1583108428133": {"attribute_name": "Publication type", "attribute_value_mlt": [{"interim": "Journal article"}]}, "item_1583159729339": {"attribute_name": "Publication date", "attribute_value": "2019-11"}, "item_1583159847033": {"attribute_name": "Identifier", "attribute_value": "https://ecor.yueco.edu.mm/handle/123456789/621"}, "item_title": "A Simulation Study on Robust Alternatives of Least Squares Regression", "item_type_id": "21", "owner": "1", "path": ["1582965639643"], "permalink_uri": "http://hdl.handle.net/20.500.12678/0000001610", "pubdate": {"attribute_name": "Deposited date", "attribute_value": "2020-03-05"}, "publish_date": "2020-03-05", "publish_status": "0", "recid": "1610", "relation": {}, "relation_version_is_last": true, "title": ["A Simulation Study on Robust Alternatives of Least Squares Regression"], "weko_shared_id": -1}
  1. Yangon University of Economics
  2. Yangon University of Economics Research Journal and Universities Research Journal

A Simulation Study on Robust Alternatives of Least Squares Regression

http://hdl.handle.net/20.500.12678/0000001610
http://hdl.handle.net/20.500.12678/0000001610
38ce81dc-79e3-470b-a28d-18d6ad2857c5
21bf05ca-21af-46e8-9bc6-cb85d69ad100
None
Preview
Name / File License Actions
Dr Dr Maw Maw Khin A Simulation study on Robus Alternatives....pdf (538 Kb)
Publication type
Journal article
Upload type
Publication
Title
Title A Simulation Study on Robust Alternatives of Least Squares Regression
Language en
Publication date 2019-11
Authors
Maw Maw Khin, Dr.
Description
Five methods of regression namely the ordinary least squares, least absolute value, M, least
median squares and least trimmed squares are applied to the multiple regression model. The
several distributional assumptions of errors are considered in this study. The required data sets are
generated by using multiple linear regression models with three explanatory variables. Then, these
data sets are transformed into outlier contaminated data sets. After that, the performances are
compared in terms of bias and mean squared errors criteria and then the most suitable estimation
method is chosen. Same sets of simulated data are used and mean squared errors and bias of these
methods are compared. It is found that ordinary least squares estimation under a heavy-tailed
distribution does not yield outlier robust estimates. Indeed, not only with the Gaussian distribution
but also with the skewed distributions, ordinary least squares estimators collapse in the presence of
small levels of outlier contamination. The Huber M-estimate and bisquare M-estimate estimate
have shown to be more appropriate alternatives to the ordinary least squares in heavy-tailed
distributions whereas the LMS estimates are better choices for skewed data. One best method
could not be suggested in all situations; however the use of more than one method of exploratory
data analysis is recommended in practice.
Keywords
Robust Estimators
Identifier https://ecor.yueco.edu.mm/handle/123456789/621
Journal articles
1
Yangon University of Economics
6
Conference papaers
Books/reports/chapters
Thesis/dissertations
Back
0
0
views
downloads
See details
Views Downloads

Versions

Ver.1 2020-03-08 15:58:05.930169
Show All versions

Share

Mendeley Twitter Facebook Print Addthis

Export

OAI-PMH
  • OAI-PMH DublinCore
Other Formats
  • JSON

Confirm


Back to MERAL


Back to MERAL