An empirical comparison of neural network and logistic regression models

Kumar, Akhil; Rao, Vithala R.; Soni, Harsh

doi:10.1007/bf00996189

Cited by 79 publications

(47 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The neural network approach produces better classification, handles complex underlying relationships better, and is stronger at interpolation [5]. ANNs have a high modelling flexibility and adaptability, as they can control the learning process.…”

Section: Neural Networkmentioning

confidence: 99%

Predicting Sales Revenue by Using Artificial Neural Network in Grocery Retailing Industry: A Case Study in Turkey

Penpece¹,

Elma²

2014

IJTEF

View full text Add to dashboard Cite

Abstract-Forecasting sales quantity and sales revenue is very vital for a company to take action for the next period for sustainable competition. It is especially important for growing industries like grocery retailing industry. Turkey's grocery retailing industry is evolving rapidly. Due to increasing importance; the aim of this study is to forecast the sales revenue of grocery retailing industry in Turkey with the help of grocery retailers marketing costs, gross profit, and its competitors' gross profit by using artificial neural network. Artificial neural networks are models which are used for forecasting because of their capabilities of pattern recognition and machine learning. ANN method is used to forecast the sales revenue of upcoming period. According to results there are high similarities between forecasted and actual data. Forecasted results of this study are bigger or smaller than the actual data for only 10%. Because of this high accuracy, companies at grocery retailing industry in Turkey can use ANN as a forecasting tool.Index Terms-Sales revenue, forecasting sales revenue, grocery retailing industry, artificial neural network.

show abstract

Section: Neural Networkmentioning

confidence: 99%

Predicting Sales Revenue by Using Artificial Neural Network in Grocery Retailing Industry: A Case Study in Turkey

Penpece¹,

Elma²

2014

IJTEF

View full text Add to dashboard Cite

show abstract

“…Neural networks have been applied to a large variety of problems going from consumer pro®ling (Kumar et al, 1995) and market segmentation (Hruschka and Natter, 1994) to market response analysis (Hruschka, 1993), database marketing and scoring (Desmet, 1995). It has been shown that ANNs can be viewed as nonlinear generalizations of models such as linear or logistic regressions (Hertz et al, 1991;Ripley, 1994), discriminant and factor analysis (Baldi and Hornik, 1989) or cluster analysis (Kohonen, 1988).…”

Section: Introductionmentioning

confidence: 99%

Neural networks and the multinomial logit for brand choice modelling: a hybrid approach

Bentz

Merunka²

2000

J. Forecast.

104

View full text Add to dashboard Cite

The study of brand choice decisions with multiple alternatives has been successfully modelled for more than a decade using the Multinomial Logit model. Recently, neural network modelling has received increasing attention and has been applied to an array of marketing problems such as market response or segmentation. We show that a Feedforward Neural Network with Softmax output units and shared weights can be viewed as a generalization of the Multinomial Logit model. The main dierence between the two approaches lies in the ability of neural networks to model non-linear preferences with few (if any) a priori assumptions about the nature of the underlying utility function, while the Multinomial Logit can suer from a speci®cation bias. Being complementary, these approaches are combined into a single framework. The neural network is used as a diagnostic and speci®cation tool for the Logit model, which will provide interpretable coecients and signi®cance statistics. The method is illustrated on an arti®cial dataset where the market is heterogeneous. We then apply the approach to panel scanner data of purchase records, using the Logit to analyse the non-linearities detected by the neural network.

show abstract

“…Applications of credit scoring have been widely used in different fields, including: accounting and finance (Altman and Narayman, 1997;Pendharkar, 2005;Landajo et al, 2007), marketing (Kumar et al, 1995;Thieme et al, 2000;Chiang et al, 2006), and general applications (Walczak and Sincich, 1999;Usha, 2005;Nikolopoulos et al, 2007). In the area of accounting and finance, credit scoring applications have been used for different purposes including: bankruptcy prediction (Zhang et al 1999;Tsai and Wu, 2008;Etemadi et al, 2009;Nanni and Lumini, 2009); scoring applications (Huang et al, 2006;Crook et al, 2007;Huang et al, 2007;Erbas and Stefanou, 2008)) and classification problems (Ong et al, 2005;Laha, 2007;Ben-David and Frank, 2009).…”

Section: Literature Reviewmentioning

confidence: 99%

Would credit scoring work for Islamic finance? A neural network approach

Abdou

Alam

Mulkeen

2014

International Journal of Islamic and Middle Eastern Finance and Management

View full text Add to dashboard Cite

Purpose -The main aim of this paper is to distinguish whether the decision making process of the Islamic financial houses in the UK can be improved through the use of credit scoring modeling techniques as opposed to the currently used judgmental approaches. Subsidiary aims are to identify how scoring models can reclassify accepted applicants who later are considered as having bad credit and how many of the rejected applicants are later considered as having good credit; and highlight significant variables that are crucial in terms of accepting and rejecting applicants which can further aid the decision making process.Design/methodology/approach -A real data-set of 487 applicants are used consisting of 336 accepted credit applications and 151 rejected credit applications make to an Islamic finance house in the UK. In order to build the proposed scoring models, the data-set is divided into training and hold-out sub-set. The training sub-set is used to build the scoring models and the hold-out sub-set is used to test the predictive capabilities of the scoring models.70 percent of the overall applicants will be used for the training sub-set and 30 percent will be used for the testing sub-set. Three statistical modeling techniques namely Discriminant Analysis (DA), Logistic Regression (LR) and Multi-layer Perceptron (MP) neural network are used to build the proposed scoring models. Findings -Our findings reveal that the LR model has the highest Correct Classification (CC) rate in the training sub-set whereas MP outperforms other techniques and has the highest CC rate in the hold-out sub-set. MP also outperforms other techniques in terms of predicting the rejected credit applications and has the lowest Misclassification Cost (MC) above other techniques. In addition, results from MP models show that monthly expenses, age and marital status are identified as the key factors affecting the decision making process. Research limitations/implications -Although our sample is small and restricted to an Islamic Finance house in the UK the results are robust. Future research could consider enlarging the sample in the UK and also internationally allowing for cultural differences to be identified. The results indicate that the scoring models can be of great benefit to Islamic finance houses in regards to their decision making processes of accepting and rejecting new credit applications and thus improve their efficiency and effectiveness. Originality/value -Our contribution is the first to apply credit scoring modeling techniques in Islamic Finance. Also in building a scoring model our application applies a different approach by using accepted and rejected credit applications instead of good and bad credit histories. This identifies opportunity costs of misclassifying credit applications as rejected.

show abstract

An empirical comparison of neural network and logistic regression models

Cited by 79 publications

References 7 publications

Predicting Sales Revenue by Using Artificial Neural Network in Grocery Retailing Industry: A Case Study in Turkey

Predicting Sales Revenue by Using Artificial Neural Network in Grocery Retailing Industry: A Case Study in Turkey

Neural networks and the multinomial logit for brand choice modelling: a hybrid approach

Would credit scoring work for Islamic finance? A neural network approach

Contact Info

Product

Resources

About