Financial Distress Prediction in the Australian Mining Industry using Tree-Based Stochastic Techniques

Khaled Halteh, Adrian Gepp, Kuldeep Kumar

Research output: Contribution to conferenceAbstractResearchpeer-review

Abstract

Financial distress is a socially and economically important problem that affects mining
companies the world over. Having the power to better understand – and hence aid mining
businesses from failing, has the potential to save not only the company, but potentially prevent
economies from sustained downturn. Mining in Australia generates around $138 billion
annually, making up more than half of total goods and services. This paper uses financial data
from hundreds of Australian companies in the mining sector to work on advancing financial
distress modelling by utilizing cutting-edge stochastic models, namely: decision trees,
stochastic gradient boosting, and random forests, to develop the most accurate technique in
forecasting insolvency risk. Our results indicate that stochastic gradient boosting was the best
technique at correctly classifying the successful and distressed companies within the mining
sector. Our model showed that net gearing, return on equity, and book value per share ratios
were found to be the variables with the best explanatory power pertaining to predicting
financial distress of mining companies.
Original languageEnglish
Publication statusPublished - Nov 2017
EventThe 29th Asian-Pacific Conference on International Accounting Issues - Kuala Lumpur, Malaysia
Duration: 5 Nov 20178 Nov 2017
Conference number: 29th

Conference

ConferenceThe 29th Asian-Pacific Conference on International Accounting Issues
CountryMalaysia
CityKuala Lumpur
Period5/11/178/11/17

Fingerprint

Prediction
Mining industry
Financial distress
Boosting
Gradient
Distress
Modeling
Decision tree
Insolvency risk
Book value
Stochastic model
Return on equity

Cite this

Halteh, K., Gepp, A., & Kumar, K. (2017). Financial Distress Prediction in the Australian Mining Industry using Tree-Based Stochastic Techniques. Abstract from The 29th Asian-Pacific Conference on International Accounting Issues , Kuala Lumpur, Malaysia.
Halteh, Khaled ; Gepp, Adrian ; Kumar, Kuldeep. / Financial Distress Prediction in the Australian Mining Industry using Tree-Based Stochastic Techniques. Abstract from The 29th Asian-Pacific Conference on International Accounting Issues , Kuala Lumpur, Malaysia.
@conference{907248ea42c94c7e974714ebd56a1952,
title = "Financial Distress Prediction in the Australian Mining Industry using Tree-Based Stochastic Techniques",
abstract = "Financial distress is a socially and economically important problem that affects miningcompanies the world over. Having the power to better understand – and hence aid miningbusinesses from failing, has the potential to save not only the company, but potentially preventeconomies from sustained downturn. Mining in Australia generates around $138 billionannually, making up more than half of total goods and services. This paper uses financial datafrom hundreds of Australian companies in the mining sector to work on advancing financialdistress modelling by utilizing cutting-edge stochastic models, namely: decision trees,stochastic gradient boosting, and random forests, to develop the most accurate technique inforecasting insolvency risk. Our results indicate that stochastic gradient boosting was the besttechnique at correctly classifying the successful and distressed companies within the miningsector. Our model showed that net gearing, return on equity, and book value per share ratioswere found to be the variables with the best explanatory power pertaining to predictingfinancial distress of mining companies.",
author = "Khaled Halteh and Adrian Gepp and Kuldeep Kumar",
year = "2017",
month = "11",
language = "English",
note = "The 29th Asian-Pacific Conference on International Accounting Issues ; Conference date: 05-11-2017 Through 08-11-2017",

}

Halteh, K, Gepp, A & Kumar, K 2017, 'Financial Distress Prediction in the Australian Mining Industry using Tree-Based Stochastic Techniques' The 29th Asian-Pacific Conference on International Accounting Issues , Kuala Lumpur, Malaysia, 5/11/17 - 8/11/17, .

Financial Distress Prediction in the Australian Mining Industry using Tree-Based Stochastic Techniques. / Halteh, Khaled; Gepp, Adrian; Kumar, Kuldeep.

2017. Abstract from The 29th Asian-Pacific Conference on International Accounting Issues , Kuala Lumpur, Malaysia.

Research output: Contribution to conferenceAbstractResearchpeer-review

TY - CONF

T1 - Financial Distress Prediction in the Australian Mining Industry using Tree-Based Stochastic Techniques

AU - Halteh, Khaled

AU - Gepp, Adrian

AU - Kumar, Kuldeep

PY - 2017/11

Y1 - 2017/11

N2 - Financial distress is a socially and economically important problem that affects miningcompanies the world over. Having the power to better understand – and hence aid miningbusinesses from failing, has the potential to save not only the company, but potentially preventeconomies from sustained downturn. Mining in Australia generates around $138 billionannually, making up more than half of total goods and services. This paper uses financial datafrom hundreds of Australian companies in the mining sector to work on advancing financialdistress modelling by utilizing cutting-edge stochastic models, namely: decision trees,stochastic gradient boosting, and random forests, to develop the most accurate technique inforecasting insolvency risk. Our results indicate that stochastic gradient boosting was the besttechnique at correctly classifying the successful and distressed companies within the miningsector. Our model showed that net gearing, return on equity, and book value per share ratioswere found to be the variables with the best explanatory power pertaining to predictingfinancial distress of mining companies.

AB - Financial distress is a socially and economically important problem that affects miningcompanies the world over. Having the power to better understand – and hence aid miningbusinesses from failing, has the potential to save not only the company, but potentially preventeconomies from sustained downturn. Mining in Australia generates around $138 billionannually, making up more than half of total goods and services. This paper uses financial datafrom hundreds of Australian companies in the mining sector to work on advancing financialdistress modelling by utilizing cutting-edge stochastic models, namely: decision trees,stochastic gradient boosting, and random forests, to develop the most accurate technique inforecasting insolvency risk. Our results indicate that stochastic gradient boosting was the besttechnique at correctly classifying the successful and distressed companies within the miningsector. Our model showed that net gearing, return on equity, and book value per share ratioswere found to be the variables with the best explanatory power pertaining to predictingfinancial distress of mining companies.

M3 - Abstract

ER -

Halteh K, Gepp A, Kumar K. Financial Distress Prediction in the Australian Mining Industry using Tree-Based Stochastic Techniques. 2017. Abstract from The 29th Asian-Pacific Conference on International Accounting Issues , Kuala Lumpur, Malaysia.